Once you’ve finished your research and analyses, you can begin working on the results section of your dissertation. This is where you present the main findings of your research and answer your research questions or test your hypotheses.
Different types of research involve a variety of tools, including observations, surveys and interviews; desk research is also a common choice. What your results section will look like depends on the kind of research you have undertaken, but you still proceed in generally the same manner.
This article provides a roadmap for presenting results related to both quantitative research (surveys) and qualitative research (interviews). Note that all of the steps below relate explicitly to your results chapter.
Results of quantitative research (surveys)
Step 1 – Introduce the relevant research
Explain that your survey was completed by a sufficient number of respondents and that you have analyzed the results.
Step 2 – Report the results in a structured manner
The easiest way to report your results is to frame them around any research sub-questions or hypotheses that you formulated before starting your research (as presented earlier in your dissertation). Your findings in relation to these sub-questions and hypotheses will together enable you to eventually answer your main research question.
Step 3 – Discuss the results
For sub-questions, briefly discuss (in one or two sentences) any results that relate to each question and make any relevant observations. However, don’t go so far as to actually answer your research questions or draw any formal conclusions (this happens only in your conclusions chapter).
The process is similar for hypotheses, although in this case you should state whether each hypothesis is actually proven by your research.
Step 4 – Add further tables or figures (optional)
You can include tables or figures (such as charts), but only if they reflect your results and add value to the story you are trying to tell. Make sure that you refer to these graphics in your running text, to ensure that readers understand how they relate to the point you are trying to make.
Note that it’s not necessary to fully explain each table or figure in the text; readers should be able to figure this out on their own. You can make it easier for them to do so by designing your tables and figures in a clear and logical manner and choosing a descriptive title.
Step 5 – Double-check your results chapter
Review the results you have addressed to make that that they all relate to your research questions/sub-questions and hypotheses. If results any are not directly related, either delete them entirely or shift them to an appendix to your dissertation.
Results section example: Write-up of a quantitative survey
The first hypothesis was tested by means of a regression analysis that used donation intention as the dependent variable and social distance as the independent variable. The results of this analysis (see Table 5) show that social distance has a significant weak link with donation intention (b * = 18, p = .05), which makes it a significant predictor of donation intention; once social distance increases, donation intention increases as well. As a result, H1 is not confirmed. To the contrary, this result suggests a significant effect in the opposite direction.
Results of qualitative research (interviews)
Step 1 – Introduce the relevant research
For example, note how many interviews you conducted and explain how you transcribed and coded the data that you collected.
Step 2 – Report the results in a structured manner
If your dissertation includes specific sub-questions or hypotheses, you can discuss the results on this basis. Answering these questions and testing your hypotheses is an important step that prepares you to answer your main question (which you will do in your conclusion section).
In qualitative research, however, interviews are not always directly related to specific questions or hypotheses. In this case, it’s better to discuss your results using the list of topics (or questionnaire) that you used for the interviews. These topics reflect the issues you are investigating and are thus tied in to your main question.
Step 3 – Discuss the results
Highlight the results that relate to each sub-question, hypothesis or topic and explain why they are relevant. You can clarify and support your observations with quotes from the interviews (so long as they speak to the topic).
You should also include a brief explanation (one or two sentences) that indicates precisely what those results mean in relation that that particular sub-question, hypothesis or topic. You might even think of this as a short conclusion – but remember that the significance of this conclusion to your research question should not be addressed until your conclusions chapter.
If you have formulated any hypotheses, you should indicate whether your research has proven or rejected them.
Step 4 – Double-check your results chapter
Take another look to ensure that all of the results you have presented relate to you sub-questions, hypotheses or topics.
Other results (such as full transcripts of your interviews, if appropriate) can be included in an appendix.
When asked what types of video games are more or less a form of art, the respondent
When asked what types of video games are more or less a form of art, the respondent replied that video games themselves are not an art form but that creativity is sometimes involved. The criteria used to identify art games consider aspects of a game that relate to its design, story and music as well as the creative teams working on it. One respondent noted the following:
“Well I think that in role-playing games [eh], more [eh] attention to [eh] character design […] to world design […] because there is a great […] because the whole story is important and more attention is paid to certain game elements […] so that perhaps […] that there is more [er][…]. Yes, […] there do you need bigger teams of creative experts than […] in […]. For example, an average shooter or something.”
It can be concluded that video games are perceived as art forms themselves, but it is clear that the creativity and art elements are more important in some games than in others.
What is the difference between the results and conclusion chapters?
The results chapter presents brief observations for each sub-question, hypothesis or topic (as seen in examples 1 and 2 above). In each case, the observations should only be one to two sentences long. Take care to not begin discussing the significance of these observations to your main research question.
This is where the conclusions step in. The results of all of the sub-questions, hypotheses or topics in the results chapter are merged in the conclusions chapter. Together they form the answer to the main research question, which is then discussed in detail in this chapter.
Table of Contents
“[I]t is a truism to note that all transcription is in some sense interpretation …” (Cook, 1990, p.12)
In the first post of this 2 part series on how to transcribe an interview for dissertation, I gave you on overview of the transcription process, 3 ways to transcribe your interviews and made a few remarks on accuracy of the transcripts and audio quality. In this second and final post in this series on how to transcribe academic interviews for dissertation I get into the minutia of transcription. How do you transcribe? What are the different ways you can transcribe your thesis interviews (with examples)?
How do you transcribe interviews for dissertation?
How do you transcribe, what do you do when you are transcribing. Well things you need to think about as you transcribe are first the names of respondents. It is useful to just use a standard format for entering the names. I suggested some of them here.
You can put the person’s actual name and perhaps the initials letter of their surnames. So for Mary Clark, you can write Mary C on your transcript. Learn to consistent, every time Mary talks you have the same letters in front of those speech. You can also identify the interviewee simply as “Interviewee” or “Respondent” or even “Resp”. Again be consistent.
For the interviewer you might have “I” for interviewer or IV, or INT or Intvr. You can put your name on it if you like, G for Gandalf and so on. But be consistent, so you use the same name all the time on the transcript to identify when the interviewer speaks. At weloty we use Intvr: and Resp: as our default speaker identifications.
You also need to think about the formatting of the speaker designation. The most common approach is have the name/identifier, and then you have the colon, and then indented text. In other cases some transcribers put the name of the speaker on a separate line. So there will be an identifier and then the next line the speech starts. You might also want to bold or italize the interviewer or the interviewee to make it easier for you to identify who is speaking. Here are some examples:
Jane= interviewer, Paul =interviewee
Tell me how you came to be a transcriber?
Well, it’s a funny story, I started transcribing when I was…
Resp: Yes, and they keep referring their fellow students to my transcription site.
Intvr: Last question, what do you find to be the most challenging part when transcribing interviews?
Resp: That’s a difficult one. Most of the time it’s the audio quality of the interview that is..
Now which format you use might depend on what software you are using. There are different standards for different kinds of software. So you think need to think about that. For instance, Atlas-ti or Nvivo require that their transcripts be formatted differently. But again read the manuals that come with the software. That there was no single best way of doing it, one program wants it done one way, one program another way. And some QDAs will accept different formats than others.
Anonymization of your transcripts is very important. As far as possible make sure that if anything is published when you do the research the name should be anonymized so that the names of people and contextual names like the organizations they work for, the towns they work in and so on, all of those are anonymized.
But you will need to keep the original names themselves. It’s common practice to keep the names on the transcripts during the analysis. So I have the original data with the original names on it. But when it is published then it gets anonymized at that stage.
On the other hand you might decide it is better anonymize it right from the start. The problem with that is you lose the context in your mind. For instance if you anonymized Mary to you respondent 2 or, June or something like that, so you have to remember June was actually Mary or respondent 2 was Mary.
Makes it slightly harder for you to think about what you are doing and what you can remember you have used there. So, that is why I prefer to keep it unanonymised until the last minute, which you will need to do that eventually. But remember to publish only anonymized versions.
Checking for Accuracy
Checking for accuracy. What if you can’t hear things? Then use a standard thing like this square brackets 3 dots square brackets […], to indicate something missing. Something was said, but none of us can work out, the typist can’t, I can’t, we can’t work out what was said, just can’t remember it. So it is missing.
Other possibilities, [bribery?] Did the person actually say bribery? I am not really sure. So put it is in square brackets put question mark, just to indicate that you are not quite sure about what is being said. In this instances it might also be really helpful to insert timestamps. So […][00:05:34] to clearly indicate where the missing part is in the audio or [bribery?] [00:08:32].
Also printing with wider margins, because you are going to be writing on things. The idea is that when you start to code and work with this material you do write on it a lot. So double spacing or a spacing and a half, between the lines and wide margins to write in on the side of the page are really helpful, because you need space to write in those notes. You don’t always have to use them all the time. But you know some bits will be very detailed and lots of lines, lots of comments on the side, so leave space for them.
Structure the transcripts. Two things I want to talk about here, 1) is to do with the software and a lot of these things depend on what software you are using. Some of it you simply do it consistently, so if you have got a structured questionnaire, where everybody gets asked the same questions, then make sure you use something like Q1, Q2 and the question perhaps, and always using exactly the same wording on everyone’s transcripts. So you can always be consistent about how do you find things, and how you search for things.
If you are going to use NVivo, put those questions headings in a heading style. Very similar to styles in Word. So NVivo uses the same style as word. So putting the Heading 1 style, when you upload the transcripts into NVivo, you could automatically code those headings as codes in the programs. Please contact us to discuss your NVivo transcription requirements, so we can decide what the best fit is for you.
Section format is another issue particularly if you are doing NVivo or Atlas-ti. Some software allow some use of automatic coding. For instance, I’ve formatted transcripts so that there were 2 kinds of returns between the speeches.
The speech was always the lawyer asking a question, and the witness replied. I had 2 returns, then the lawyer asking the questions, then a single return, then the witness replying and then 2 returns. I did that because the software the client was using automatically assign each of those speakers interchanges as a paragraph.
So again it needs reading the computer manuals to check, what is going on and how you can then set things up for what you need.
Styles of Transcription
I talked a little about this in Part 1 of this series, but let’s have a deeper discussion about the levels of transcription. The problem is people don’t speak in whole sentences. They repeat themselves, they hesitate, they stutter, they talk in very long sentences. There is no full stops in speech. They use contractions like don’t, coz, I’m and so on. And they use filler words as they hesitate, you know, I mean, err, mmh and so on, all sorts of different sounds.
The question is do you transcribe all those things that people do? Do you spend time with all these issues, repetitions, hesitations and so on? Now for some purposes you do do that. If you are doing conversation analysis then this is the kind of detail you need to transcribe.
So the question is how much of the interview interaction do we want to capture? I think there are 3 different ways of transcribing: what I call the intelligent verbatim approach, the (strict) verbatim approach, and finally Discourse or the Conversation Analysis, CA. Deciding which approach to use largely depends on your research questions and the intent of the study. As an important step in data management and analysis, the process of thesis transcription must be congruent with the methodological design and theoretical underpinnings of each investigation.
Let me show you some examples of each of these approaches.
Intelligent Verbatim Approach
To begin with the intelligent verbatim approach. Now, I plan to pen a detailed guide on intelligent verbatim transcription so keep your eyes peeled out for it.
Intelligent Verbatim Transcript Example.
Frank: True, it’s going to be the greatest in the history of the world.
Jon: I’d expect no less. I think it will put an end to the conflict in the Middle East, I mean quite possibly.
Frank: That’s true.
Jon: No, no I’m confusing that with a nuclear bomb. I’m sorry, yours is not going to bomb. Yours is actually going to do really well. But let’s share some ideas I want to run by you and then if I can get some information from you it’s going to help a whole lot.
This is snippet from a interview transcript between Jon and Frank. The purpose of the interview was to gather information on a product that’s being launched. So the intelligent verbatim approach works really well in this particular case. It reads fairly easily and from the transcript you can quickly get the information you need. This type of transcription is perfect for researchers using the grounded approach or analyzing for themes and categories.
Verbatim Transcription Approach
I have written a couple of detailed posts on verbatim transcription. The first is a general overview of verbatim transcription and the other a detailed verbatim transcription guide for psychotherapy interviews and sessions. Please refer to those posts for a more detailed guide on how to create verbatim transcripts. A short example.
Verbatim Transcription Example.
Frank: Oh true, it’s going to be the greatest in the history of the world.
Jon: I I I would expect no less. I think it will put an end to um to the conflict in the Middle East I mean quite possibly.
Frank: [CT] That’s true.
Jon: No, no way I’m I’m I’m confusing that with a nuclear bomb. I’m sorry, that’s that’s, yours is not going to bomb. Yours is actually gonna do really well. But ah [chuckles] let’s ah let’s share some, I guess I guess some ideas I wanna run by you and then if we can talk through um, if I can get some um information from you it’s gonna help a whole lot.
Same interview, different transcript. As you can see, there’s more detail of the interview interaction in the verbatim transcript as opposed to the intelligent verbatim transcript. Verbatim transcripts are great if you interested in the dynamics of the interview.
Discourse or the Conversation Analysis
Conversation analysis (CA) is a “unique” (ten HAVE 1990) form of qualitative social research. Books have been written about this approach to transcription = there is no enough space on this blog for me to adequately cover conversational transcription convections. But, I’ll provide you with a few references that you can use. Jefferson, Gail (1985). ‘An exercise in the transcription and analysis of laughter’ is a great starting point and a valuable entry point into learning more about the Jeffersonian CA transcription notation. I’d also recommend Ochs, Elinor (1979) ‘Transcription as theory’. Even for us non-discourse analysis researchers, it’s a great read. And for a critique: Ashmore and Reed (2000) Innocence and Nostalgia in Conversation Analysis.
Discourse Analysis Transcript Example.
Here is an image of a discourse analysis transcript. There are many symbols used conversational analysis transcription and creating an appendix of what each symbol denotes is the first thing you do before you begin transcription.
If you are going to use a conversation analysis approach you do need this kind of level of transcription of the data. But for most of us this is not relevant, even if you are doing discourse analysis, this is not relevant. Verbatim approach should get you what you want.
Dissertation Transcript Cover Page
As you transcribe interviews it is useful to have with them information about the interview itself, or about the case if there are separate interviews perhaps. So it is common to have a document headers sheet with the data. Again if you are doing it in software, then you keep it in a certain place. In NVivo document properties is a place where you keep this information.
Typically things you would have here are, choose the names. If you anonymize the interviewee’s name, use this coversheet or document header sheet keep the pseudonym, you can separate it after the interview if you want to.
You can also add information about location, time, topic and circumstance, and add your interview notes. Also include the name of the interview, if you are in a team of people it might be useful to name who is interviewing the person.
If you are taking field notes as well, and I certainly would advice that if you are doing interviews to take notes as well recording the interview, you take notes about important things that you think about, or things that happen, or ideas that occur to you as you interview the people. So again link it or have the name of the document on the cover sheet.
Here’s a sample dissertation interview cover page you can use.
Interviewee: [Name of interviewee] [Pseudonym]
Interviewer: [Name of interviewer]
Date and Time: [mm/dd/yyyy][00:00]
Location: [Place interview was conducted]
Audio file information: [Name][Duration]
Link to field notes:
Link to follow up interview transcript:
That’s it on this series on how to transcribe your interviews for dissertation. If you have any burning questions post them below and I’ll be more than happy to answer them. And if you find transcribing your dissertation interviews to be a chore – get in touch. We’ll be glad to transcribe them for you.
Ashmore, Malcolm & Reed, Darren (2000). Innocence and Nostalgia in Conversation Analysis: The Dynamic Relations of Tape and Transcript . Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 1(3), Art. 3.
Cook, Guy (1990). Transcribing infinity: Problems of context presentation. Journal of Pragmatics, 14, 1-24.
Have, Paul ten (1990). Methodological issues in conversation analysis. Bulletin de Méthodologie Sociologique, 27 (June), 23-51.
Jefferson, Gail (1985). An exercise in the transcription and analysis of laughter. In T. Van Dijk (Ed.), Handbook of Discourse Analysis, Vol. 3: Discourse and Dialogue (pp.25-34). London, UK: Academic Press.
Ochs, Elinor (1979). Transcription as theory. In E. Ochs & B. Schieffelin (Eds.), Developmental Pragmatics. New York: Academic Press.