A methodological story, or how I planned to create a (metaphorical) chest of drawers and ended up with a bedside table
The story of the project: collecting the data (cont)
I began a preliminary analysis of the 24 interviews I had by then conducted by listening many times to the recordings I had made on a digital voice recorder and by writing a summary of each one. Never having had any experience of conducting research interviews, I had been very nervous at the outset that the technology would function as required, although I had practised using it several times. Unfortunately, within several minutes of the start of my first interview, with Tony (a pseudonym), a group of porters entered the room we were using and proceeded to move out all the furniture – noisily – except for the chairs we were sitting on and the table that the recorder was balanced on. For fear of stopping and starting the recorder and finding out later that it had not recorded Tony’s narrative fully, I continued with the interview and Tony soldiered bravely on through the noise. This was not the most auspicious start for a novice interviewer – fortunately, when the recording was uploaded onto my computer the sound was not affected and the summary was written up like all the others, although I always laughed to hear the background noise of chatter among the porters and the screech of chair legs across the wooden floor. I also worried that the disruption had affected Tony’s responses but felt that the second interview I conducted with him six months later would give him the opportunity to give a more reflective account and counteract any distortions. While I attempted to find a quiet place for the subsequent interviews, which were all held in institutional settings, several more involved interruptions. The difference was that I now trusted my technology and could stop and start the interviews smoothly, when required.
A feature of many of the interviews I conducted for this project was that they did not begin and end when the digital recorder was switched on and off. In the case of some of the managers, who had given me a particular time-slot for the interview, there was just time for a brief preamble and my explanations of the project often took place as part of the recorded interview itself. In contrast, with many of the German lecturer respondents I often had a lengthy conversation before switching on the recorder and even more so once the recorder had been switched off. This was partly due to the stance I had adopted towards the practice of conducting life history interviews. I had considered using the biographical narrative interpretative method (BNIM) espoused by Chamberlayne (Chamberlayne et al 2000, 2002) and Wengraf (2006) but rejected it because it required a collective approach to the interpretation of the interviews based on grounded theory, and I was a lone researcher, and because it tended to emphasise psychological rather than sociological explanations (Wengraf 2006). What I did like about the method was the open structure it allowed for in a first interview, something also supported by Goodson (2005), who advises not to prepare for life history interviews too much. By not supplying questions in advance, as some respondents had requested in order to prepare themselves for our interview, I was able, as a reflexive researcher, to take part consciously in the construction of the life history narratives of my German lecturer interviewees as they took place.
The recorded interviews typically began with an open question, such as ‘tell me about how you came to be a German lecturer and how the recent changes have affected your perception of your professional identity’ and proceeded with questions by me as prompts to the respondents to expand in greater detail on the issues surrounding their professional experiences. After reviewing the first six respondent interviews I realised that I had not always got a clear definition from each person regarding how exactly they defined their professional identity – I therefore emailed them to ask how they described what they did and received several responses, some of which I quoted in my analysis, as appropriate.
I was always conscious that I did not want to cause difficulties or provoke anxiety for my respondents – some were clearly editing their narratives while the recording was going on and were much more revealing of their feelings, and often frustrations, when the recording had stopped. Because of the ethical duty I owed to these individuals I decided not to include most of these asides (except for a few anonymous examples to illustrate the power relations between the interviewees and myself which I will discuss below). I saw the summaries of our interviews as a form of contract – in the sense that ‘this is the information I will include, not anything else that may have transpired between us’. When I later sent each of the interviewees a copy of their individual interview summary I invited them to comment or correct any misapprehensions on my part. A few respondents replied to correct factual misunderstandings and the HR manager asked me to amend any impression that he was commenting on events that took place before he had been employed at his institution. The duration of the interviews ranged from about 22 to 52 minutes and ended when I felt that no new material was being produced and we had reached saturation point on the issue of changing professional identity and the events that engendered it. The summaries were on average two to three pages in length and included occasional direct quotations from the interviewees, wherever these appeared particularly expressive or pithy.
For the second round of interviews with nine of the ten respondents from one IoT this time the interviews were more structured and directed towards answering questions that sought to map the lecturers’ accounts onto Bourdieu’s concepts according to a matrix that I had devised as part of the analysis of the first round of interviews. (The matrix and analysis will be discussed in the next section.) I also prepared questions specific to each interviewee in order to explicate gaps in their career trajectories that seemed apparent from the first interview. Conscious of the intrusion (Bourdieu et al 1999)a> of returning a second time to people who had busy professional lives I said from the outset that the interviews would be short (about 15 minutes) and that I would provide a full transcript in order to show where I would be drawing the bulk of their quotes from in the thesis. At this point I feared that some respondents might withdraw from the project if they felt that they had been too open and critical – fortunately for me, no one did. I emailed similar questions based on the matrix to the six German lecturers from the other institutions but did not contact the three former part-time lecturers or the five informants for any further information.
Copies of the interview transcripts were sent out with a covering letter which drew out one further discussion point from each interview and invited the respondents to comment on my analysis of this point. Two respondents commented briefly on the points I had raised and another wrote an extensive and reflective response on the evening after our interview. Two more respondents, whom I met several months after the second round of interviews, passed general comments about the transcripts but did not put these in writing.
This, then, was the extent of my primary data; the boundaries had been fixed and the next stage was to work with what I had been given in order to provide a meaningful analysis of the changes my respondents had made in their professional lives and how these changes were affecting their professional identities. Secondary data sources included institutional documents, such as quality assurance and review documentation, minutes of meetings and institutional policy and position papers, as well as information and policy documentation from Irish academic staff organisations, the IoT sector and the national and European higher education policy fields.