Overall, this interview exercise has shown that there are many issues that need to be kept in mind when conducting interviews. Factors to take into account can be divided into three groups:
1. Pre-interview preparation;
2. During the interview;
3. Post-interview reflections.
In order to prepare for the interview, Michael and I discussed the overall research objective/aim of the study in order to situate the study in the relevant context. We decided that the study would be on the experience of international students studying at ANU. Next, we felt that we had to develop an interview protocol with open-ended interview questions. This provided some structure for the semi-structured interviews so that they were all to some extent similar and comparable as the same questions were asked to each interviewee. Developing suitable interview questions took some time as we decomposed our overall aim into parts to generate specific interview questions. Given that this is a qualitative study, I have noticed that when writing questions certain terms or words can be used that are more appropriate for these kinds of studies. For example, some words commonly used in interview questions for qualitative research include: perceptions, experiences, describe, factors, influences, affect, attitudes, etc. It is important to bear in mind that the order of questions as well as the questions themselves may lead to possible response bias, and therefore, it is the responsibility of the researchers to minimise this bias. When constructing our interview questions we tried to keep them as neutral as possible so as to reduce potential interviewer bias. For example, our first question asked how they would describe their experience as an international student at ANU in general. This left open the possibility for interviewees to mention any aspect that they considered relevant to them.
We selected subjects at random and who were not acquainted with the researchers and we found this process to be quite effective as the subjects asked all consented to an interview. In practice, when selecting case organizations for case studies and potential interviewees it may help to develop contacts/connections, know key informants and have some kind of relationship with the organization. However, I wonder if there is a difference in potential responses if researchers know or do not know the interviewees personally. This may be something that could be investigated further.
Conducting the interviews was a relatively smooth process overall as we were well prepared going into the interviews. I think that the researchers need to be prepared going into an interview in order to make it flow as smooth as possible. For instance, this may include making sure that both the researchers and interviewees are fluent in the language in which the interview is conducted in, interviewers being familiar with the interview questions and broad objective of the study, making sure that your recording equipment works, checking the environment (e.g. the room) in which the interview will be conducted in, developing some expectations of possible answers to check whether responses confirm or contradict your expectations.
Another issue that has come to mind is whether you provide consenting interviewees a list of the interview questions before an interview. In this way, they have some way of preparing for the interview and can already come up with some data or responses to give. Because sometimes when you are asked a question on the spur of the moment you might not come up with all the answers on the spot. However, if you have some time to think over the interview questions, you my come up with richer answers during the interview. Moreover, you may also provide some company documents that help answer some questions.
During our interviews with international students we found that interviewees, at times, did not fully understand our questions. Based on this observation I see that a major issue when doing interviews is making sure that your interviews understand and do not misinterpret your questions, otherwise the responses may not help you in collecting relevant data. I found that having two interviews conducting one interview is very useful as they can take different roles as well as switch during an interview. For example, our plan was for one of us to recruit an international student, one of us interviewed them as the other person observed and took observational notes, made sure the recording equipment worked and also listened to the interview to check that all the salient points had been covered and mentioned any other issues. Because you do not realise it at first but there are many tasks that an interviewer needs to complete during an interview and it helps to have two people alternating roles.
Another observation is that sometimes you need to let the interviewees talk if they want to talk as they may reveal some important information in their stories. We conducted three interviews with three international students and I found that one interviewee in particular talked the most and, in turn, provided us the richest data. However, at the same time, often it is even what is not said that is equally important. Some responses for some questions were a little short as there appeared to be a bit of a language barrier.
Body language and external environmental factors are important issues to take note of when conducting interviews. I found that it is not just the words said but also the way they are said, the tone of the voice and the disposition of the interviewee that all provide meaning and context, and therefore help with interpretation. For example, one interviewee seemed a little rushed during the responses, however they still provided us with important data. You need to consider if there are any distractions and whether the interviewee is comfortable. Our interviews were conducted in three different locations - a quiet room, in the foyer of the library and outside. We noticed that our three interviewees had three distinct profiles which reflected in their responses provided.
After the interviews, transcribing them took a while and we consulted useful references to help code and analyse our raw data. So in qualitative research there needs to be some kind of systematic approach to provide meaning and interpretation to your data. I think that qualitative research has some element of grounded theory in it, particularly if you do more interpretive research. I look forward to learning more about grounded theory in qualitative research because it seems to some extent quite different and radical to the more scientific method of research.
In transcribing our interviews in verbatim, we noted that there are other items that are often included in transcripts, such as ..., ahhh, uhmm, etc. In analysing the data we used established coding frameworks, and worked on one question at a time. First, decomposing it into discrete parts, then comparing these parts across the three interviewees to identify similarities and differences, and lastly, establishing potential connections and relationships amongst the responses. This process helped us to identify common or unique emergent themes in the data.
Lastly, after the interviews I took some reflective and analytical notes and memos in order to jot down some pertinent points and observations of the interviews. Here I reflected on my observations and and what I thought rather than analysing the data from the transcripts. I found that this exercise helped we when I later had to code and analyse the data from the transcripts. So, in a sense, when conducting interviews the interviewer has multiple tasks and roles. Also, you obtain different sources of data, the responses to the interview questions from the interviewees as well as reflective and analytical memos of the interviewer.
No comments:
Post a Comment