I’ve certainly been thinking through these issues myself now that I’ve started transcribing interviews for my own doctoral research. In the end, I have had to think about how I will use the transcriptions.

For the most part, I do not need to do a detailed conversation analysis, and I’m finding it easier to annotate rather than “perfectly” transcribe. In some cases, a few notes about a section rather than a transcription are all I need (saving a lot of time). Because I am doing my transcription on the computer I can always come back and add more detail to the text if I need.

No text is ever a complete rendition of all that is said and unsaid. Indeed, no audio is a complete record of all that occurred in the interview or conversation. Something is always left out, and open to later (re)-interpretation. That’s why my own annotations are often more useful than carefully marked up text. It’s only ever going to be about my analysis of the event, and since I am producing my own text in response to the interviews, not a court document, I might as well start the analysis as I transcribe.