The first pilot interview occurred with a colleague, John, at the University and this was significantly unsuccessful. Not because of the interviewee but because of the nature of the subject I wanted to explore and the way I constructed the interview process.
I wanted to limit bias to the discussion and sought to enact an informant-led interview. Such an interview is characterized by limited intervention. As a consequence there were only three questions:
- How would you define leadership and describe your leadership practice?
- What are the key incidents that have influenced your thinking about leadership?
- What rules of thumb guide your leadership practice?
I expected these questions would allow expansive discussion. This was far from the case. The comments on the first and last question were lucid but seemed disconnected. The middle question was very difficult for John and the discussion lacked depth.
He could not clearly recall much that had influenced his thinking. In light of this disappointing interview, the second interview with a senior colleague was much more structured with many more questions to provoke discussion in the areas I wanted to explore, that is, how have you learnt to lead?
The second interviewee described a feeling of frustration that he was inhibited in discussing his experience in a more integrated manner and he felt that the structured set of questions did not enable exploration.
Reflecting on a Michael Parkinson’ television interview, it struck me that the interview was structured chronologically; Parkinson worked his way through the person’s life relating the discussion to pre-interview chronological notes. In the spirit of good science I experimented on myself.
It seemed useful – I discovered some interesting insights that I had not reflected on previously. The process seemed to order and structure my thinking. It put together my experiences into a holistic story. I read work on narrative methodologies and found that I had hit on a rich vein of knowledge that would guide the development of my naive embryonic timeline process. Particularly useful was the work of Kuhnert and Russell (1990) and Wengraf (2001).