I stumbled across this article “doing qualitative research as if counsel is hiding in the closet”, a very good example of how a researcher proceed when dealing with research ethics regarding confidentiality and inform consent in such particular studies as “political elites”. When providing a description of the research before the beginning of the interview, researcher hand the “ethical review board” or IRB-approved information sheet about the research and tell them that “nothing they say will be directly attributable to them”. In journalist´s parlance, the interview is “off the record”.

Bellow I have pasted the most significant part:

A lot of my research studies political elites. As such, I am frequently conducting participant observation and open-ended, semi-structured interviews in the halls of the US Congress, offices of various federal agencies, political consultants, lobbying firms, and organizations that aim to represent the public’s interest. In other words, my data are collected from a highly educated group of people, an overwhelming proportion of whom have law degrees. Moreover, most of these offices employ some sort of “corporate” counsel that monitors access—or what I think of as my field site and my research subjects. As a result, I have learned to be extremely careful since these lawyers have made it clear to me on a number of occasions that I can lose access and be booted from my field site at any point.

In the 15 years since I completed my PhD, I have been challenged by research subjects regarding my use of their names or the data I collected from them in two particularly anxiety-inducing cases. In the first, a subject of an interview who worked for a Congressional Committee found a draft of a paper online that directly quoted him. While I was making the final edits on my first book, which named this subject and quoted him directly, I got a very aggressive email from him. In response, I passed on a copy of the transcript of the interview that included an exchange during which I asked if I could use the subject’s name and he affirmed. His concerns were alleviated after receiving the transcript that included his consent. Nonetheless, I removed direct reference to this research subject in my book. I also adapted the way that I approach political elites whom I study.

Although these interviews are usual seen as exempt from IRB requirements because I am asking about subjects’ political work and not anything personal, I have found I get better data (and avoid such interactions with JDs working in the political arena) if I grant all subjects confidentiality. When providing a description of my research before I begin an interview, I hand my subjects an IRB-approved information sheet about the research and tell them that nothing they say will be directly attributable to them. In journalists’ parlance, the interview is “off the record.” I state that I will email them directly for approval if I find there is any segment that I would like to quote directly in my work. Because so many of my subjects have experience speaking with journalists, I find that following similar norms about attribution puts the subjects at ease. Although this process adds some work when I am writing, it tends to yield more interesting data. This process also gives me an electronic trail if I am approached by subjects post hoc, which can be very helpful if I am contacted by lawyers.