- The bowtie
- Thesis introduction:. This is what I want to know: The question – and any sub-questions
- Literature review: Why I’m doing this: The literature (right bow)
- Method: How I’m doing this: The method: evidence/data collection (the knot)
- Findings/results: This is what I found: The findings/results (left bow)
- Discussion: Why does it matter? The discussion – Tying the bowtie
- Abstract: The last thing you write – the clasp
There are various ways to compile data into a thesis. Statistically-based work may be short and sweet with copious appendices of measures and statistical data; narrative studies lengthy with multiple chapters containing verbatim transcripts or recordings that present the findings by theme or perspective. Thesis supervisors and committees may expect differing degrees of depth in the literature reviewed and in the discussion of findings. Most theses however share two key qualities.
First they address a question or questions that have the potential to add a new piece of information to the existing literature. The reader needs you to situate the work in the broader literature from which it is drawn, and to relate the findings back to that literature after they have been presented in order to show how the evidence makes its contribution.
Second the thesis presents data, evidence or information that answers the question or questions– and the evidence is the key voice in answering the question(s). The thesis author makes a case at the outset for why the question(s) must be asked, and what he or she claims (speculates, hypothesizes) will emerge. The evidence that is then presented provides the answers to the question(s).
In the final part of the thesis, the author draws conclusions from the findings revealed in the data, and ties them back to the initial questions and location in the literature. Because the purpose is to address one or more central questions by presenting evidence, the method and results are the core of the work.
The elements of a thesis are connected. I’ve used the analogy of a bowtie for the structure of a thesis. The ribbons that connect the bowtie at the back are like the introduction and discussion of the study. The thesis comes full circle in relating back to where it started – with the discussion of the findings reflecting back to the original questions and their place in the literature. More about the discussion later.
Here are the elements:
- “This is what we want to know” introduces the question(s) and sub-questions; the first ribbon
- “Why I’m doing this” describes how the thesis question is located in the literature; the literature review. It’s the first part of the bow.
- “How I’m doing this” – The method: evidence/data collection forms the main knot – tying the questions and their context to a method that will answer them.
- “This is what I found” – the results, evidence, the material that addresses the initial question(s). It’s the other part of the bow.
- “Why does it matter?” – the author discusses the results, relating them back to the initial question and the literature context from which it sprang, thus completing the circle with the final ribbon and clasp.
“This is what I want to know: The question – and any sub-questions”
In this section state why the problem is important and the strategy you will follow to investigate it. This is not the Abstract – more about that at the end. This is a summary to give the reader a heads up on what they can expect to find when they get beyond the literature review.
See Unit 6 for an extended description of how to write a literature review.
“Why I’m doing this: The literature (right bow)”
This is the chapter usually termed the Literature Review. Its purpose is to locate your study in the larger field; to describe the approach that others have taken in exploring this field, why is it important from a theoretical and applied perspective, the main issues that have been tackled by previous researchers and the questions that still need to be addressed . Finally you describe how your study adds to the field and why you have chosen the methodology you will present.
The style of the literature review should continue on the path described in Unit 1; non-evaluative and objective. It is not an exercise in persuading the reader to agree, but setting out the evidence that currently exists in a factual and non judgmental way. You can certainly report others’ controversial and evaluative statements, with full citation and preferably verbatim quotes. This may lead you to justify your specific research questions, but do not tip your hand about what you will conclude in the final sections of the thesis. Lead the reader to understand all perspectives of the issues you tackle, without the use of judgmental statements such as superlatives or emotional appeal.
Also the literature you present will not be exhaustive. It presents a context for the reader to locate your study. It is helpful to assume that your readers are familiar with the overall field of your study but they need to be reminded of the major trends in the larger field and introduced to any specific studies that relate to yours. So the analogy for the literature review is a funnel. It starts broad and possibly with historical trends, and becomes more detailed as you focus on the context on which your work will be based.
“How I’m doing this: The method: evidence/data collection (the knot)”
In unit 5 we discuss how to ‘operationalize’ your research question; that is, how to turn it into a series of measures, evidence, artifacts and/or transcripts that allow the reader to judge that the resulting evidence that you will present is valid and reliable. The purpose of the method is to allow the reader to decide if the evidence is trustworthy, and not the byproduct of one or another sources of bias.
Bias can occur in each of the sections of the method:
Participants. Your sample of participants should match the objectives that you set out to achieve, If this is a case study, then the sample size might be one, but the breadth and depth of that case will address the research questions you pose. As noted before a case study has strengths in contributing layers of depth and nuance to the investigation, but it cannot address comparative questions such as efficacy of a treatment, or trends in the field. On the other hand, a large scale sample of a broader population of participants might be chosen to address questions of efficacy of an intervention or treatment, a trend in the population (think electoral polls), or the characteristics of one group compared to another.
Again, sample bias can affect the interpretation of results, and the reader will want to be assured that the chosen participant sample satisfactorily represents the potential answers to the research questions. In this section you can also present preliminary characteristics of your participants, sometimes in tabular form. Particularly in a qualitative study, if you have multiple participants with a variety of characteristics, a summary table is helpful to the reader. The reader will want to refer back to such a table while reading the results.
Design. This might encompass measures and/or apparatus used to conduct the study.
In a quantitative study, the design may be experimental, quasi experimental, or mixed. The reader will be looking for the basis on which your statistical analysis was built and whether it is likely to be convincing in its validity and reliability. In statistical analyses, the reader will interpret the confidence estimates and degrees of error for this purpose.
In a qualitative study, the reader is not looking for the confidence estimates, but is certainly seeking the extent to which he or she can be confident that the evidence is true. This usually means extensive data collection to provide first-hand evidence, followed by some form of authentication by the source. For example, transcribed interviews may be returned to the interviewee for them to scan and verify as representing what they intended to say. They are then used verbatim as evidence in the thesis.
Note that any steps you take to verify the measures, details of the outcome of reliability estimates, field trials and/or peer reviews can be presented here, or at the beginning of the results chapters.
Procedure. The standard for the procedure is that another researcher is given sufficient details that he or she could replicate your study with other participants. This will be a detailed description of the way the study was conducted, its time lines and the instructions given to participants. Either in text or in appendices, you supply the wording of the invitation to participate, the ethical certificate, what the participant was told was the purpose and process, verbatim details of the instructions given to the participants, the actual interview questions or the original questionnaire drafts. Then present the way you approached data analysis; the software programs you used for analysis or transcription, and the techniques you followed to analyse the data.
Note that if you are preparing a scaled down version of your work for publication or presentation these details are usually summarized and your thesis cited as the source. But for the doctoral thesis completion, they need to appear, in order for the reader to evaluate the veracity of the findings that you will next present.
An example: In one thesis the author described the widespread resistance of the participants to a new policy that had been recently introduced. Reading the actual questionnaire that was added in an Appendix, however, I discovered that the wording was so slanted that a participant would have been unable to express any alternative opinion about the policy. So the questionnaire contained bias that inevitably led to the reported findings; a non-starter as a defensible thesis.
“This is what I found: The findings/results (left bow)”
This is the heart of the thesis and its decorative bow.
You may want to present the findings in several sections. I generally recommend starting with the ‘housekeeping’ items; the responses to your request for participants and the resulting sample (if not already in the Participants section), the statistics on your measures, the outcome of the validation process for your questionnaires or your results of pilot tests of your interview protocols, etc.
Use headings that reflected the Table of Contents at the beginning of the thesis, you then move to the major findings.
Your task is to guide the reader through your results structure. Good writing practices include setting up signposts in the form of headings that reflect the contents of the section, referring the reader to appendices that contain raw data etc. If it is a large and complex thesis, remind the reader of where links or tables and figures were last encountered or where they will be taken up. This will help the reader to navigate. Your thesis is a communication and helping the reader to access it will be part of successful writing.
“Why does it matter? The discussion – Tying the bowtie”
This section of a thesis can be difficult. On the one hand you have set out all the evidence and the reasons why they are important, and since you are thoroughly immersed in your thesis, the point that you have reached feels self evident. On the other hand, you don’t want to merely summarize again what you did and what you found.
Take a break before tackling this final – and important – chapter. It may start with a summary, but like the bowtie, it must now put the evidence back into the context from which it came. This is where you can, in a limited way, interpret your findings and bring in some of your opinions. This is where “I thought this would happen” can be stated.
“The last thing you write – the clasp”
This will be the last thing you write. It isn’t easy. It requires you to step back and, in very few words, summarize the most important aspects of your thesis. Since it is the first piece of writing that most of your readers encounter, it needs to set up your thesis and their acceptance of it, so it needs to be concise, accurate, and brief, but also to point to the context and importance of your study.
The APA Manual of Style suggests how to do this in 120 words, the requirement for a scholarly publication. Most academic institutions allow longer abstracts; – check the requirements for your specific university, if they exist. Otherwise aim for about 350 words.
It’s the clasp of the bowtie – it draws the whole thing together into a brief summary.
Finally, take time to attend to the details. Check that the citations in text appear in the list of references and vice-versa. Check that tables and figures appear in the sequence listed in the Table of Contents. Inevitably someone on your committee is going to go through your draft with a fine toothed comb and find typos and errors. Don’t worry. It’s standard procedure!