We collected a lot of data through Geran PdP and FRGS. Herewith the method we used to analyze those semi-structured interview transcriptions in a group:
Step 1: Align research questions with the list of interview questions
Step 2: Sort raw data (verbatim interview transcription) using tables based on research questions and sub-questions (which are interview questions)
Intended outcome: List of direct quotes (in tables)
Step 3: Interpret each quote using description, by following the sequence of research question
Intended outcome: List of interpretations (in parallel to the original quotes)
Step 4: Construct mind maps to sort, synthesize and link relevant quote / interpretation
Intended outcome: Research question oriented mind maps
Step 5: Generate conclusive statements as answers to each research question
Intended outcome: List of conclusive statements and list of research answers
NB: Since not all research questions are fully answered, answers generated from analysis of raw data might not be sufficient to make conclusion of this study, thus two different lists should be presentation.
Thanks for your information in Interview Transcription. Hope devotes will be careful after reading this post.Regards.
ReplyDeleteInterview Transcription