Approximately 70 semi-structured interviews with people participating in the ACIA process were conducted. This included meetings with participants in the Assessment Integration Team, lead authors, and representatives of national delegations to the Arctic Council. A list of interviewees is presented in Appendix II. Some people were interviewed several times and, in total, the group that I talked to at one time or another included 56 people. The purpose of the interviews was to gather information that was not likely to have been elaborated in written documents and to gather information on the participants' own analysis of the process. As Stake expresses, the interview is the main road to multiple realities.19 Interviews can thus be a way to facilitate triangulation, especially in establishing a description of events. As noted previously in the discussion on observations, interviews were also used to probe some issues noted during observations. In addition, interviews gave participants in the ACIA process an opportunity to discuss the process from their particular perspective in a way that they may not have been able to do in the social setting of a formal meeting. Moreover, keeping in line with the view of interviews as conversations and a situation in which the interviewer and the interviewee co-create a description, interviews were also used to gather reactions on my own observations and preliminary interpretations. 20
The co-creation of description and analysis with interviewees entails the risk of imposing a picture on the person interviewed rather than letting him or her speak their own mind. To reduce the introduction of such potential bias into the material, the interviews generally started by asking open-ended questions that urged people to describe events or processes as they saw them. For example, lead authors were asked questions regarding the task for the chapter for which they were responsible and/or what types and sources of information had been most important in writing the chapter. Interviewees were generally well articulated speakers. My perception was that they were likely to let me know if they did not share a suggested interpretation thus lessening the risk of introducing bias into the observations.
One group of interviewees were selected based on whether they have a central role in the ACIA process (e.g. chair of the Assessment Steering Committee and chair of the Policy Drafting Team), whether they are especially well informed on specific issues (e.g. the liaison between specific chapters and the Assessment Integration Team), and if they presented a unique perspective on some part of the process (e.g. a specific country representative or indigenous peoples' representative). These interviews generally did not use a specific interview guide, but relied more on giving interviewees a chance to elaborate either on specific issues or developments during the ACIA process or to elaborate on their perceptions of the process as a whole. Most of these interviews were done in connection with formal meetings, however eleven interviews were done over the phone.
The second group of interviewees were the ACIA lead authors, with whom I specifically discussed the content and process for their respective chapters according to an interview guide, see Appendix III. The purpose was to get the scientists' own perspec
19 Stake, The Art of Case Study Research, 64.
20 Steinar Kvale, InterViews (Thousand Oaks: Sage Publishers, 1996).
tive on what knowledge had been integrated in the ACIA process and why. Other questions focused on how participating in this activity had affected their own picture of climate change in the Arctic and their perceptions of science-policy interactions. Such semi-structured interviews were carried out with 23 of the 26 lead authors. All lead authors were contacted via e-mail to arrange the interviews. Most of the interviews were conducted in connection with the ACIA Scientific Conference in Reykjavik in November 2004. Two people, who were not at the conference, were interviewed via the phone. Follow-up attempts were made to cover all chapters of ACIA's scientific report, but one chapter is unfortunately lacking in interview material.21
The lead author interviews ranged from a little more than half an hour to over an hour in time. Circumstances varied but many of the interviews were carried out in a lobby where other activities were going on and where we were sometimes interrupted. Some of the other interviews were more time pressed. In a few cases we deliberately sought out places to talk where we would not be disturbed.
The basic agreement with lead authors was that the interview was "on-the-record," which would allow me to quote by name. However, to give them a chance to discuss sensitive issues with a promise of confidentiality, they also had the opportunity to add comments "off-the-record," which several of interviewees used for specific comments. For other participants in the process, the nature of the interview and the way I was allowed to use the material was generally agreed upon at the beginning of the conversation. For many participants in the policy process, the agreement was that I could use their comments as background and list their name in a general list of people whom I had talked to but not to quote them by name.
All interviews were recorded with a digital tape recorder and most of them were transcribed. As I did transcriptions myself, the transcription process went hand in hand with the analysis.
A major advantage of interviews as a method in this study was that they provided a window into parts of the ACIA process that were neither formally documented nor accessible for observations. Without the interviews, it would have been impossible to analyze ACIA's policy process. Capturing the dynamics of producing the scientific and overview documents would also have been very difficult otherwise. The interviews also provided an opportunity to build trust with participants in the ACIA process, without which it would have been difficult to provide an account that the participants could see as legitimate. As part of this trust building, an effort was made to interview people who might have different perspectives on contentious issues, e.g. representatives from different member states and Permanent Participants in the policy process.
Interviews as method also have weaknesses, including introducing bias by leading questions or being vulnerable to the interviewee's bias or poor recall of events.22 The major way to guard against interviewee bias in description of events has been to rely on more than one interview or if possible corroborate accounts with written documentation.
21 Chapter 12 is not covered despite repeated attempts to reach the lead author. Of the other two lead authors who were not interviewed, one deferred to his co-lead and the other did not respond to the initial request. Because the chapter was already covered by two lead authors, no follow-up attempt was made.
22 Yin, Case Study Research. Design and Methods, 86.
Interviewer bias is more difficult to control, but as discussed previously, interviewees' senior academic or political positions made it less likely that they would place themselves in an accommodating role in relation to me as an interviewer. Several of the interviewees also had experience in dealing with the media.
Was this article helpful?