I am finally ready to start recruiting participants for my doctoral thesis on Identity Positioning of Doctoral Students in Networked Learning Environments. The seemingly endless wait for ethics permission and institutional permission (sometimes known as site permission) was actually a very useful time. I used the time for further reading and reflection on phenomenography in light of the recent methodology workshops and my readings on methods.
Bowden, J. A., & Walsh, E. (2000). Phenomenography (p. 154). Melbourne, Australia: RMIT Publishing. Retrieved from http://www.informit.com.au/products/ProductDetails.aspx?id=PHENOMENOGRAPHY_ERIN
Marton, F., & Booth, S. (1997). Learning and awareness. Mahwah, NJ: Lawrence Erlbaum Associates.
Why study interviewing methods so intensively right now?
I feel strongly that one must understand one’s methodology before collecting data. And, it is important to be mindful of all the choices that one must make while employing data collection methods. Some researchers utilize data collected previously by other researchers for other research purposes. This might be suitable avenue for researchers who have difficulty accessing some participants or whose projects are on tight time constraints. It might also be done simply for convenience. In some cases, seemingly perfect data sets can be procured from national databases such as the UK Data Archive. As I read more deeply about phenomenography, I sense that it is in the best interests of my own study to collect data carefully with sensitivity to phenomenographic theory and procedures of analysis. This is also the position taken by Bowden (2000):
Whatever research method is used, researchers need to articulate the purposes of their project and to keep those purposes in mind at all stages of the research—in the design of the investigation, in the development of the data collection processes, in the collection of data and in the analysis and interpretation. (p. 7)
What is a phenomenographic interview?
Interviewing is the most common method for collecting data in phenomenography (Walsh, 2000, p. 19, Marton, 1986, p. 42). Trigwell (2000) and Dunkin (2000) suggest that the ideal number of interviews rests around 15 to 20. In phenomenological studies, the number of suggested interviews might be less as phenomenology seeks to explore the essence of an individual’s experience in some depth. Phenomenography, on the other hand, focuses on the limited possible ways of experiencing a given phenomenon across a group of individuals. The outcome space (results) is a compilation of categories of description which expresses the variation. As such, it is necessary to reach a balance between depth of description and breadth of experience among a group of individuals.
Interviews are typically semi-structured (or “guided” if you choose Olson’s terminology) and last roughly 40 to 60 minutes or until the “the interviewer feels the experience has been described, and the meaning of relevant words has been revealed (Trigwell, 2000, p. 67). Most phenomenographers seem to agree that the participants should have sufficient flexibility to describe the experiences as they wish in their own way. Hence, most questioning and probing is open-ended. Bowden (Chapter 1) and Prosser (Chapter 3), in working with children, prefer to offer “problem questions” that the participants are asked to resolve. By asking participants to work through problems or tasks “interviewees are encouraged to reveal, through discussion, their ways of understanding a phenomenon, that is, to disclose their relationship to the phenomenon under consideration” (p. 9). Most of the questions in a phenomenographic interview follow from comments of the participant (Trigwell, 2000). Some sample questions that Bowden supplies (p. 10):
- Could you explain further?
- What do you mean by that?
- Is there anything else you would like to say about this problem?
Prosser offers some interesting observations on eliciting information from the participants:
While it is relatively easy to get interviewees to describe their strategies, it is much more difficult to get them to discuss their intentions underlying their strategies and their conceptions of phenomena. (Prosser, 2000, p. 44).
To approach the participant’s conceptions, the researcher must consider the interview/discussion as a whole. According to Marton & Booth (1997) interviews take place on two levels: the interpersonal contact between the interviewer and the participant and at a metacognitive level in which the participant relates his/her awareness of an experience (Marton & Booth, 1997, p. 87). So, whilst the researcher attempts to maintain focus on the target conception(s), he/she must also provide room for the participant fully express related nuances and details. In some cases, it might be helpful within the context of the co-constructed interaction (see previous blog postings on interviews), for the researcher to share her own experiences. However, Bowden (Chapter 4) warns against “leading too much” to avoid influencing the participants. As per my previous blog postings on interviewing, anything that the researcher discloses during the interview should be as carefully transcribed and reported as the participant’s dialogue.
In some cases, a phenomenographic interview might seem to revolve tediously around the same question over and over again. This is partially true. A phenomenographic interviewer will ask similar questions in different ways so as to elicit a number of different views on the phenomenon. “Typically, a range of questions is used to provide views of each conception from several angles in order to make the description of the conception as rich as possible” (Dall’Alba, 2000, p. 94).