This was never my dream, but it has almost been my entire thesis up until recently. In this post, I will describe what my interviews have been about, and how I have structured my transcription.
What did my data consist of?
The raw interviews consists of 22 audio files, on average about 30 minutes long. The longest interview was 46 minutes, and the shortest one was only 19 minutes – in total 11 hours of recording. Each interview consisted of two parts of content: A qualitative first part assessing general characteristics of the participant and also preparing the participant for the second part of the interview which was directly related to the research questions.
Transcriptions were done manually in Word using VLC to play the audio files. VLC allows you to play the file at a slower speed (60-80% of normal), pause and play using keyboard shortcuts, and define ‘jumps’ in the file manually (2-4 sec jumps).
A good transcription method is consistent in order to make the transcription reliable. My rules were primarily based off of this Guide to Transcribing by Áine Humble. It was sufficient to simply make appropriate adaptations to fit my project. I divided my rules into the following topics: General remarks; General formatting; Pauses, inaudible material, or interruptions; Nonverbal communication; Confidentiality; Quoting; Miscellaneous. Overall, my transcription guide ended up containing quite specific rules that fit my interview purpose. A couple of examples:
- I use “P” for participant and “I” for interviewer. Each comment is typed single spaced with a blank line between each comment:
I: I’m starting the interview now
I: Let me first say thank you so much for coming.
P: You’re welcome.
- If a person interrupts themselves mid-sentence (trails off or doesn’t finish it, or they change their thoughts part way through a sentence), a long dash afterwards (an em dash) will indicate so. Example:
P: The acceleration goes – the force is aligned with the movement
P: The acceleration goes, the force is aligned with the movement
P: There’s no acc – acceleration in the direction of the movement
P: There’s no acc, acceleration in the direction of the movement
- As much nonverbal communication as possible is identified in the transcripts (pauses, laughter, crying, gesturing, sighing, and so on) using square brackets and the identified communication in present participle. For example:
P: I’m not sure I understand why I choose this response [laughing].
- Interpretation of the nonverbal communication is omitted. For example:
P: I’m not sure I understand why I choose this response [nervous laughing].
- For numbers less than 10, the number is written (e.g. “seven” instead of “7”).
Final notes on my transcription process
I found it a lot more challenging to transcribe interviews, and it took much more time than expected. My future process will be to continue my analysis based on these first transcriptions and later on listen to the recordings again and make minor corrections to them. This will inevitably affect my final results a bit (but not much), and will in turn allow me to continue my work much faster.