Ross Woods, rev. 2018-25
In many institutions, students hand in their work to their supervisors part by part, often in chapters. You will normally need to have edited each chapter into good enough shape so that the supervisor will want to read it. While it doesn't need to be perfect, it needs to be free of grammmatical mistakes, typographical errors, and layout glitches. Research supervisors generally want to discuss the research aspect of your dissertation and don't want to talk about annoying writing errors. In fact, if they find lots of writing errors, they are quite entitled to return your draft and ask you to make those corrections before they will even read it.
However, the main part of the editing is the last stage before submission, where you will read and re-read the whole dissertation many times.
When writing a first draft, your main task is to get everything into words on the screen. Many advisors say that you shouldn’t edit at all during this stage, although I don’t mind it a little. The problem with editing at this stage is that it is easy to get distracted by imperfections and then have difficulty getting the first draft written.
When you are satisfied, put it away for at least three days, and preferably a week. This will give you the opportunity to look at it with fresh eyes. Then re-read it. You will probably see small errors that you missed before. (It is difficult to check your own writing, because you read what you think you meant or what you remember, not what is actually written.) You might find yourself saying Why did I say that?
, What did I mean by that?
, or That's not what I meant to say.
or This seems jumbled and confusing.
This gives you a chance to fix these errors while you can still remember enough and before anybody else sees them.
Note: Keep backups, because computer data can be lost very easily.
For more information on editing, see the separate link.
How will you store data securely and maintain confidentiality?
Quantitatve data normally needs cleaning. This involves deleting duplicates and correcting any spelling errors that would affect computer processing.
Does anything need translation?
Do you need to transcribe interviews of voice recordings in preparation for analysis, such as interviews done in qualitative research. You can pay a transcription service to do it for you, but some students do their own. Although time-consuming, it gives more opportunity to understand the data well, and might spark insights and analyses that would otherwise more difficult. If you are writing in English and your time is more valuable than the fees for commercial software applications, you can try nvivo, temi.com, rev.com, Trint, or Otter. Google Chrome also has free plugins for transcription. Here’s a tip from Tοm Granοff. Zoom webinars has an automatic transcription feature for its cloud recordings. The transcription can be exported into Notepad, then into your word processor, and then presumably into your Qualitative Data Analysis package. See https://www.youtube.com/watch?v=3dk7xk3WtzE.
The analyis is basically the bridge between your data and your findings.
Collate and arrange your data in a neat, legible, systematic layout. This will require interpreting the data, and the way you do it will depend on your methodology and the kind of data.
Think about your data and how it will solve your research problem and answer your research questions. Draw conclusions, and check that they relate to the main research problem as most recently defined. This is the most abstract step of all, and perhaps hardest to define. How it's done will depend on your specific topic and the kind of data you are using. (The importance of alignment will become immediately clear; the method must produce data that will enable you to solve the original research problem.)
If data is extremely consistent, your research might get onto the fast-track. However, your research could get very interesting if your data is diverse or inconsistent. You need to find out why that is so. Perhaps your method was good, but other particular factors caused inconsistencies. It might also have been aspects of your methodology that could not have been anticipated during planning.
Findings are the results directly derived from the data analysis. For example, a researcher might report that Variable A correlated closely with variable B.
The conclusion is your answer to the research question that you proposed in the introduction.
In qualitative research, these tend to be the same thing; the results of the data analysis are expressed as the answer to the research question.
In quantitative research, however, they can be different and separate. The results are the results of mathematical analysis. while the conclusion is an inference drawn from the results.
❓ How can I know that the data will answer my research question?
It will if you have gathered data that addresses the research question. This is why alignment is so valuable.
However, the answer you get might not be the answer you anticipate. Some students even think that their data is wrong if it leads to conclusions that they didn't expect.
❓ Can a conclusion be a summary statement of the research?
It depends. In some cases of qualitative research, it can if it answers the research question. In other cases, it is simply going around in a circle and missing the point, so that there is no real conclusion.
❓ Can a conclusion have multiple elements?
Yes, some research is complex. However, it would be easy to list various elements and not actually have a conclusion. In other words, one really needs a one-sentence statement, even if it is qualified and expanded in various ways.
This is a general procedure for interviews, focus groups, program evaluations, and ethnography. It is not necessarily suitable for other kinds of qualitative methods.
When to stop collecting data.)
distinct tendency to ...
All respondents gave answers consistent with ...
Responses were quite polarized ...
As you work on your analysis:
In a dissertation, conclusions and findings can be placed in a section at the end of the analysis chapter or in the Conclusion chapter.
💡 Get a colleague to check your coding and structure. This will make your results more objective and help you to avoid inserting your personal attitudes or opinions into the analysis.
⚠ Some mistakes are easy to make if you make incorrect assumptions about your respondents:
confirmation bias.)
the shaman effect.)
❓ How good does my data have to be?
The data must have enough rich detail to lead to confirmed conclusions, and lots of analysis will not make a poor data-set adequate. Your data is still inadequate if most occurences are only short comments with little information. In that case, you might still need to gather more data, that is, hold more interviews or focus groups to get more detail.
❓ I've read all my transcripts many times, and I think I don't have enough data. What should I do?
You probably have enough data if you followed the instructions in sampling and When to stop collecting data
. Your methods means that collecting more data won't be helpful.
❓ What if I can't see a pattern?
When to stop collecting data, then they probably have enough data.)
This is what my data is doing ...and then you describe the pattern in your data. The next step is to put it in writing and edit it so that it is concise and accurate. If that doesn't help, speak to your supervisor.
In a dissertation, conclusions and findings can be placed in a section at the end of the analysis chapter or in the Conclusion chapter.
As you work on your analysis:
The method chapter should be very easy to write:
Unfortunatey, if you didn't follow your plan and kept inadequate notes as you went, you will have to do a lot of difficult work to reconstruct your method accurately, and might be at risk of failing the program.
Method chapters also have some strict rules:
In some institutions or departments, the discussion is part of the analysis chapter. In others, however, it is a separate chapter, and this is a probably a little easier for students, because it makes it into a separate stage of writing with a separate purpose. In some institutions, it is called Implications.
Despite little consensus as to what precisely should be in it, there is considerable agreement that it should present at least the immediate implications of the findings. In this view, its purpose is to compare the results of the current research with other related research.
⚠ Make sure your discussion does not undermine your research or bring it into doubt.
The most common and straightforward way to explore implications is to compare your results with those of other researches on similar or related topics. This kind of comparison demonstrates how you have added to knowledge on the topic, and indicates progress. (This assumes that your research is closely related to exisiting literature and that you are adding to a body of knowledge that is to some extent monolithic.)
When comparing your findings with the studies in your literature review, did your work:
Most or all of these other sources should be in your literature review. However, be careful not to write another literature review; this discussion of the literature should be different because it has a different purpose.
Examine what would or should change as a result of your research. Your study might also have other interesting and important wider implications, and this is also the place to explore them. In this section, and only here, you may wander off track a little because implications often go wider than the specific research topic. If you have not done so in the discussion, you should ask, How should existing theories be modified?
How should existing practices be modified?
Other questions might have arisen. Are other bodies of literature now relevant? If so, you could mention them briefly. Have you identified new significant parallels with other fields of study? If so, can you ask about wider theoretical principles or implications that need to be explored?
In a thesis or dissertation, the conclusion is a separate chapter. It indicates the finishing point of your journey, so it needs to give your readers closure. It shows that you argued for a particular conclusion and recaps on how you've reached it. The core of the conclusion (the thesis for which you have argued, and which is the answer to the main research question) should be expressed in a single sentence.
*Writing the closing paragraph and the last sentence is usually very difficult. They should usually give a general application. However, they should not be so broad as to be vague or meaningless, and you may not introduce new elements that were not discussed earlier.
What future research do you suggest? For example:
residuein your conclusion. These are fairly normal in descriptive linguistics, and comprise items left over after analysis; they are unexplained exceptions to normal patterns in the language.
When possible, tell a story; put the information in chronological order and make it flow. It will engage your readers and make reading more enjoyable. It will also demonstrate your point, which is equivalent to providing evidence to prove your point. However, be very cautious of bias, which an easy trap if you choose a particular perspective from which to present your story.
If you are working with a particular theory, you will normally state it in your introduction, either as an assumption or as a theoretical framework. During the dissertation, keep in discussion
with the theory. You might find that your evidence sometimes agrees with the theory and sometimes does not. In doing so, you are contributing to the theory, by either supporting it or modifying it.
You should update your literature review at this stage. It is probably a while since you first wrote it, and new publications since then might have changed the face of your topic. However you handle it, you will need to include them in your literature review, so that your dissertation is up to date when submitted.
Some theses and dissertations don't need any appendices, although one institution requires all students to include the ethical clearance as Appendix A.
In certain cases they are necessary; they normally include information or documents that are necessary to the research but would disrupt the reader if included in the main text. The two main categories are raw data and ancillary documents.
If you need to know whether to include something as an appendix, consider these questions. If in doubt, discuss them with your supervisor:
• Is it necessary to the research?
• Would it disrupt the reader if included in the main text?
Write the introduction, which will be Chapter 1. The introduction gives the reader a clear direction on where the whole thesis or dissertation is going, explains the topic, and gives information that readers will need to interpret it correctly.
Most of the introduction has already been written, but you still need to edit it, so that you can put in it any assumptions and new directions that come up during research. You can also add any delimitations, definitions, and assumptions that you identified since writing the proposal. (You can also clarify assumptions where they are most pertinent in the body of the text.)
The parameters of the work need to be clear, even if they are implicit. Poorly defined boundaries have the potential to cast doubt on the conclusion. Parameters may be defined in terms of specific populations, programs, movements, periods, bodies of literature, etc. Sometimes parameters are purely arbitrary, but you should explain what they are and that they are arbitrary.
Beware. Do not give definitions for ordinary words which give no cause for concern. Include definitions only for terms that are specific to the field of study and are either ambiguous, vague, or used inconsistently. (You might find competing definitions of some key terms in the literature.)
You already have a working title, but now you need a final title. Do this last to make sure you get it exactly right. As you write the dissertation, you might have better understoond what it is actually about, so your final title might differ from your original working title.
Write the preliminaries. At the very least, you need a title page and a table of contents. Most institutions also require an approval page and an abstract. See the style guide for more assistance, for example, lists of tables and figures. Do not give a dedication.
Your software might automatically generate the table of contents and lists of tables and figures. However, you will still need to check the formatting.
Include a preface if your personal experience is necessary to understand the content of the dissertation.
The main purpose of an abstract is to inform other researchers what you did and what you found in order for them to know whether they need a full copy for their own research purposes. Consequently, it must usually be able to function as an independent document with its own identifiers and explanation of any acronyms used. Some abstracts are published as separate documents. Either way, abstracts should be written only after the body of the text has been approved. They must usually have Abstract
at the top as the main title.
Institutions and publishers of abstracts usually set limits in word totals. Simply follow the rules of your institution. Consider these variations:
Abstracts increasingly follow a fixed format comprising the following parts:
A common wording for abstracts is as follows:
The purpose of this study is ...
The scope of this study ....
The methodology ...
The findings ...
Conclusions reached are ...
Limitations of this study include ...
This study contributes ...
Another common outline is known as IMRaD, meaning Introduction, Methods, Results, and Discussion
:
Introduction: What are your reasons for doing this study?
Methods: What methods did you use? Decribe them briefly.
Results: What did you find? Give the answer to your research question, as well as any other secondary findings.
Discussion: How do the results compare to previous research? What contribution have you made? What are the implications?
Collate all the parts and edit it into one harmonious document. You should now have the following:
* Includes adjustments arising from implementation, including explanations.
** Findings and conclusions can be different. The findings in a quantitative study might be the results of the statistical analysis, while the conclusion is a statement addressing the research question or hypothesis in the Introduction.
***Where should you put long verbatims? Unless your supervisor allows otherwise, view the full text of verbatims as raw material and put them in appendices. The text should predominantly contain your description and analysis supported by brief, pertinent quotations from the verbatims. The post-modern trend, however, is put long verbatims in the main text. The rationale has been that they are valid research because they let the speakers speak, without the researcher’s interpretation and editorializing.
dumpedinto a detailed discusson without being first informed about its purpose.)
This outline is not obligatory. For example:
chapterscan be so long in a large dissertation that they can beome multiple chapters. For example:
With thanks to Dαwιe Vαn Vuuνen
The standard for dissertation presentation and nearly the same that of a published book, and academic standards are generally based on publishers’ standards. In fact, one of the purposes of a dissertation program is for the student to demonstrate that he/she could write an original academic book. This is no longer simply a standard; most dissertations in accredited institutions are now published, even if non-commercially in a research repository on the internet.
You might feel that re-writing and editing are a frustrating waste of time, but an essential part of the task is polishing the manuscript and gettting the details right. This is the stage to get meticulous. Get details in logic and terminology correct, and make sure your spelling, punctuation and references are as close to perfect as you can make them. It is a basic academic skill, and is almost the same as writing a book for a publisher. According to one source, theses and dissertations are frequently rejected for small errors that are overlooked at the editing and proofreading stage. (Lyn Wαlden, 2017.)
It is comforting to know that software can now do much of the editing for you, even though it cannot do everything.
Unless you notice a serious mistake, it is now too late to add any new discovery, late data, or interesting discussion topic to yur proposal or dissertation. You really have to go with what you have, and can only make minor changes.
Theses and dissertations have two main kinds of readership:
Supervisors and assessors generally want to sit down and enjoy reading something that is pleasant and easy to read, informative, and interesting. They don't want reading to feel like hard, grinding work. Enjoyable dissertations generally have the following characteristics:
Editing usually comprises at least five different kinds of activities, and software can help with some of them:
By the end of this stage, the logic of the whole dissertation should be sound. It has to make sense to someone who reads it afresh, who has a good background on the field of study, but in not your particular topic.
The primary concern is the flow, structure, and logic. Edit these at this early stage because there is little value in perfecting the details of a large section that you might later delete.
You might need to improve the structure, for example, shrinking long-winded sections and expanding sections that are too brief. Don't worry if it reduces the page count; it's progress even if it doesn't feel like (Students often fear that deletions will reduce the word total and give assessors a bad impression.) For each thing that needs to be condensed or cut, you will probably have to add more than you delete to fill gaps that you hadn't seen before. You can learn something through each of these changes so that the final product is better.
For example:
This is editing at the level of writing. Simple, direct, concise
is still good advice; even academics want to read something engaging and interesting.
You might have noticed that some of the language in dissertations is quite complex. The question is then, “How complex should the language be?” The answer is, “Only as complex as the topic requires.” At times you might need to be complex because you use specialist terminology and deal with difficult topics or necessary details. But that is no excuse for convoluted writing that is unpleasant to read. It is still best to use simple, clear, concise writing. Do not use lengthy verbiage to conceal a lack of real thought or to increase the word total. Instead, use good language to make your point more clearly and engage your readers.
Let software do some of the checking for you. This will save you lots of time and effort, and will pick up errors that you simply didn't see. You'll normally find that the software is imperfect, so you should be willing to overrule its suggestions at times:
Your dissertation might be looking good, but you need to go through it looking for errors in grammar, punctuation, and layout. This calls for attention to detail and this is the stage to get meticulous. Get details in logic and terminology correct, and make sure your spelling, punctuation and references are as close to perfect as you can make them.
Have someone read the whole manuscript and give you honest feedback. Tell them what you want them to do, because some expect to be only a proof-reader rather than a general critic.
Readers will be annoyed if they keep finding small mistakes in an otherwise good dissertation. Go through looking for the little errors you've missed so far. These might be typographical errors that the spellchecker missed, grammar that’s not quite right, punctuation errors, and clumsy sentences.
ghost writers.
First, you need to get it done soon if a deadline is closing in on you. Second, it depends on the reader’s expertise and attention to detail. A careless reader achieves little, but some people are quite accurate and still quite fast.
For the most part, the goal of a writer is to have something that is a quick, pleasant read where your readers feel, I learned something new and helpful.
Consequently, proofreaders need not be so conscientious that they get bogged down with details. They don't need to let the task take forever or become a stressful burden. Proof-readers are not co-authors and it's not their role to find solutions to problems. They only have to tell you if they see something wrong.
In most cases, a dissertation is in bad shape if reading it is a slow and laborious task. The exceptions are those parts where the writer's priority is to plod through lengthy (a.k.a. boring) details and get them right.
When should you stop? The purpose of the final editing is to get the details right, but don't let perfectionism or lack of confidence prevent you from graduating. It is always easy to do one more edit or to look for one more journal article but you have to draw the line somewhere. It is better to have a dissertation that is not completely perfect and to graduate, than to have a good dissertation that never gets submitted.
A good dissertation is a done dissertation.It’s a compromise between being sloppy and being perfectionist; your finished version can be very good without being the final word on perfection.
When you are ready, run your paper through turnitin.com and show your supervisor the output.
If your supervisor agrees, you can now print off the final version and submit it to the committee. For fully online institutions, you'll more likely submit a pdf software version.
The oral examination will follow the procedure published in your institution's handbook, and you should familiarize yourself with it. Procedures for oral examinations, however, vary greatly between institutions and some call oral examinations oral defense
, while others use the term viva voce, which is Latin for live voice
.
Is an oral examination required?
If an oral examination is required, the arrangement depends on the institution:
Institutions also vary greatly in the resources you may use. The traditional standard is that you may have only a copy of your dissertation. Some online institutions also allow you have your textbooks with you, which is very useful if you want to quote anything on methodology.
Your supervisors will only schedule an oral defense because they think you are ready. You already know what you need to know, so be encouraged. It is almost unknown for candidates who reach this stage to fail and not get a degree at all. (It would reflect badly on the supervisor to approve a defense for a student who would not pass.)
Ask your supervisor how long in advance the committee members will need to read the whole dissertation before the final defense. It will probably be at least 2-4 weeks, but they might also want to think it over for a little longer.
If you are required to have an oral examination, your objective is to successfully defend your dissertation. You should be very familiar with all contents of your dissertation, and should be able to find any sections or topics without undue delay and ready to field any questions on them. The oral defense should be easier if you are accustomed to discussing your research. You might already have considerable experience in these skills if you are experienced in giving presentations at formal conferences and responding to colleagues' questions.
Some supervisors even conduct a mini-defense with their students beforehand to give confidence and accustom them to the process; in the best cases, students have no surprises later on during the actual defense.
If you can, attend oral defenses or ask people about their experiences. Some institutions record them, and you might be able to listen to or watch the recordings.
If you did a quantitative study, check your statistics system. Most doctoral candidates have done only one course in statistics, and the statistician is the dark horse in an examination committee. You shouldn't really have to answer questions like: Why didn't you use the ABC statistical method instead?
(mentioning an obscure method that you have never heard of), and the moderator should disallow it. But you must be able to explain the system you used, show that it was fit for purpose, and show that it produced reliable results.
More (Link opens new window.)
Be confident and talk as the expert. Make sure your methods and results discussion are solid.
Spend the minimum time recapping your proposal and the majority of your time explaining your findings.
Examiners will ask about things that you didn’t include in your presentation, although some of these might already be listed in your methodology chapter or in your suggestions for further research. Try to anticipate what they might be, such as:
“What alternative methodologies did you consider, and what effects might they have had?”
“Did you experience any setbacks during data collection?”
“Does your research have any particular gaps or limitations?”
“In hindsight, what would you have done differently?”
Examiners will probably also ask some questions you hadn't expected. For example, they might pick out what seems to be subtle weaknesses, gaps, or contradictions in your work. Don't let it fluster you; just give honest and fair answers.
The committee might require some minor corrections. It is quite normal, so just accept it and do not be discouraged. If you are required to re-submit with corrections, there should be a deadline given in the assessment report. The committee should also give you a written list of exactly what corrections they require; ask for it if they don't. (You don't want to be in a situation where they say that you didn't make all required corrections.)
Assessment methods for dissertations vary greatly beween institutions and between countries:
British programs tend to use this system of outcomes:
You do not need to give a gift of any kind to your supervisory committee, and some institutions have rules prohibiting it. However, it is quite acceptable to give your chair and perhaps committee members a bound copy. You might also include a note of thanks.
After all approvals, you should normally provide extra copies of the dissertation to major stakeholders. For example, you might provide a bound, hard copy to the major funding body, or simply a link to the soft copy.
You might also want to send an Executive Summary of findings to your study participants. It might have only one or two pages. Besides your identifying information (title your name, date, degree, and institution) it can have the following sections: background, significance, analysis, key findings, implications, conclusion. It needs to be in plain English, but otherwise can be quite similar to the abstract. You might also send a link to the published dissertation.
With thanks to Hεlεn Hicks.