Interviewing
Last week I conducted my first interview with one of my participants and have since transcribed it and begun the process of analysis. Given I have more interviews to come, I think it's helpful to reflect on how the first one transpired.
Firstly, approaching the interview felt quite difficult, not in the sense of having to be the interviewer, but in being completely objective and receptive to whatever is said whilst also having a topic that you want the person to talk about. With qualitative research, interview questions are open-ended and designed to allow the participant discuss their experience in depth, which is excellent for gathering data, but it can be hard not to ask leading questions when you're trying to get them talking about a specific topic.
I found a lot of what was said to not be fully relevant, which was to be expected when using an interview structure that allows the participant to take it in their own direction, and tried to allow this to happen and then bring it back to the wider topic.
Another thing that became apparent as I was interviewing my participant was that I felt tempted to ask certain questions based on the knowledge I had of the participant thanks to our personal relationship - that is, as the participant's friend, I knew things about her that an outsider wouldn't necessarily know.
As such, there were questions that I wanted to ask but felt may actually bias the research data. For example, I knew the participant had danced with her husband for a number of years, and while this information wouldn't in itself affect the data - in fact it was able to generate usable date - some of the particular questions about this in the frame of my topic of storytelling seemed unsuitable to ask in that they led, implied, or assumed certain experiences that may not have been brought up in data otherwise.
I think all that is to say, when you interview someone you know and are friends with about a topic that interests you for academic research inquiry, if you're not careful it's easy to get de-railed and lost in the conversation, which could result in biased, inaccurate, or poor data for the actual research.
The interview did, however, generate usable data within our conversation and so the next step was to transcribe it. I considered using software to transcribe as it is time consuming, however I decided I would do it myself as it would force me to really pay attention to the words said and I know that I personally learn a lot when I write things down, so this felt like a more beneficial (if more tedious) option.
My prediction turned out to be correct as I found that as I went through the transcription process, I came across points where I thought, 'I should have asked more about that' or 'I wish I'd asked her to expand on this' - these were things that I didn't pick up on in the moment of the interview and didn't see as markers of interest; it was only reflecting on these sentiments that made me realise the opportunities I had missed.
Unsurprisingly, this was very frustrating, especially when there is a lot of data that isn't useful to you (even if it's interesting conversation, you have to be honest about whether it has any relevance to your inquiry question) and you can see your own flaws as the interviewer and researcher.
However, it does serve to raise more questions, both in term of the analysis of that particular data and whether it can be triangulated with data in literature or my own practice. This is an ongoing process and will later be compared to data from other interviews.
The phenomenological analysis approach that I am taking is undoubtedly time consuming as it requires going through the data several times, then finding units of meaning, awarding significance to the relevant meaning units, then interpreting these individually and consequently in the context of the whole data. The meaning of the whole data and its units of significance are analysed in oscillation, each unearthing more meaning in the other.
Phenomenology also acknowledges and looks through the lens of my experiences and potential biases in the research, so although objectivity in the research process is important, the impossibility of complete objectivity is addressed by explicitly accounting for the fact that analysis is impacted by the research individual tasked with interpretation of meaning - me.
This analysis is much easier on paper than it is to actually carry out; the theory makes sense but the execution is no small task. However, this is all a learning process, and learning from the process of researching is as important as leaning as a result of researching.
As I have said previously, my two questions to keep me grounded in the analysis process in relationship to my inquiry question are:
- Why am I talking about this?
- How will my practice improve?
It's so interesting to hear you reflect on your interview Roanne, I had so many of the same feelings as you did. In my first interview, I found that my participant really enjoyed changing the topic. Although this was frustrating because I felt like my questions weren't being answered, it was really refreshing to hear all her stories, especially as she has so much experience. I have been transcribing myself too, and even though it takes me roughly 4 hours to write up each interview, it is so worth it. I find myself wishing I had asked follow up questions at the time, but I like what you have said about it gives you something to search for in literature. It is interesting to hear about your analysis approach too, and I love your two questions - I have noted them down and will be using them myself! Thank you so much for sharing :) xx
ReplyDeleteHi Alice, I'm glad you found some of your own experience in mine, I find that's always reassuring to hear when we are doing something that is new to us! The two questions are a bit of a culmination of Sam's advice so I'm sticking very closely to them! I'm glad you took something from it :) xx
Delete