Transcription: before analyzing the interviews conducted during research, it is best to transcribe them, that is, to listen to the recording of the interview (when the respondent has consented to being recorded) and put the contents in writing. When no recording is available, organize notes by respecting as faithfully as possible the spirit of the interview.
Choose among several transcription methods according to the research objective. In all cases, it is important not to select or dismiss unexpected responses, but rather to reflect as faithfully as possible the full content of the interview. You can choose between:
- Full Transcript (verbatim): write everything down, include hesitations, repetitions and grammatical errors. Also include laughter and the mood. This level of transcription is not always required. It takes longer and is more tedious to produce, but is particularly useful when analyzing speech or vocabulary, and when focusing on interactions (between the interviewer and female/male respondents, or between participants in a focus group).
- Edited Full Transcript: write everything down, but exclude hesitations, repetitions and laughter.
- Reformulated Transcript: faithfully transcribe the content, but in written text format. Therefore, do not include hesitations, repetitions or laughter, and correct grammatical errors to make the transcript easier to read.
- Summarized Transcription: do not transcribe all the content, only the key points. This level of transcription can be useful for research involving a high number of interviews where only a few specific points need to be taken into account. However, there is a risk of dismissing elements, which at first glance may seem unimportant, but turn out to be important in light of new information obtained further ahead in the research.
It is important to type transcripts quickly after conducting the interviews in order to fully understand the context in which the information was given. It is also important to take note of the respondent’s profile (ex. elder woman, raised in the bush, grandmother, etc.).
[1] Jean-Pierre Deslauriers
Analysis: once the data collection phase is over, how best to analyze the transcribed interviews, notes, review of books and articles, informal discussions and observations? It is not uncommon to feel a bit overwhelmed by the sheer number of transcript pages and notes. But it is these documents, rich in data, that will allow you to answer the research questions. Proceed methodologically to avoid getting lost and to successfully bring out the various elements in the responses. This phase is called data analysis. “The intent of analysis is to produce an explicative synthesis of the information gathered.”
Analysis involves classifying, indexing, summarizing and categorizing information. This places value, in an organized manner, on the words spoken by respondents. It is important that those words be included in the final report. A good analysis will make presenting the findings and writing the report much easier
Analysis throughout the research project can help you look out for unexpected elements that can emerge and even lead to changes in research questions or to more interviews being added to make up for missing information. Be sure to take note of these kinds of assessments and decisions in your research journal.
But eventually, the time will come to take stock of the information gathered and systematically analyze it: Are there trends? Are there divergences? What can be discerned from the compiled testimonies and notes?
One way to organize and classify information is to segment it into topics and sub-topics helpful for analysis. Some topics may have already been chosen back when defining the research topic and problem. Nevertheless, more will be added both along the way and while reading interviews because the information gathered and the testimonies from respondents will modify and build upon those initial choices.
Each researcher will develop her own method to organize and classify information. In general, paragraphs or sentences from transcripts that refer to the chosen topic are identified, either with notes in the margins or colour coding (by highlighting the relevant passages). You might then decide to group together passages on the same topic to provide a portrait of what was said regarding it, thereby facilitating analysis.
For instance, if one of the topics is land, include everything in the interviews that refers to land, and without bias: “Our land is not only the community, it is much more.” “I feel good when I’m in the community, it is my place.” “For me, it is in the in the bush that I feel at home.” And so on. By grouping together all these references to land, one analysis can be: For some people, land is primarily the community as a place for identification, for others it lies outside the community, in the bush.
Next reflect on how the data answers the initial question or leads to new questions. To return to the example above, you could conclude that the relationship to land varies for different individuals and communities. It is often wise to take a step back before drawing conclusions from analysis to give your reflection time to mature and deepen.
Saturation: the gathering of information ceases when repetitions in the respondents’ testimonies occur and when a range of different opinions has been covered. In short, saturation occurs when there are no new responses or factors in the data that could modify analysis.
It then becomes a question of building an understanding of reality, of the topic being studied. Starting from a research question, data provides the elements to frame an answer. The answer is not always the one expected. The research findings could be troubling and force you to question the ideas and opinions you had at the beginning. This is far from being a failure! Indeed, this is the reason research is such an asset!
Validation: once analysis has progressed far enough and some conclusions can be drawn, it is time for validation with research participants. This process not only serves to verify if the information was fully understood, it is also an opportunity to share the preliminary results with the people involved.
The format of validation is often a meeting with participants. In some cases, you may decide to meet all participants together, while in other cases, the meetings will occur on an individual basis. The advantage of a group meeting is that it devotes time to sharing and coming face to face with the information. The validation meeting is also an opportunity to obtain missing information from interviews. Finally, it is an opportunity for participants to also analyze and draw conclusions.
