Complementing the level of analysis.
Simply put, the level of detail included in a transcript should match the level of analysis being applied. Ethnographic work, for example, is interested in understanding values, beliefs and attitudes and how they are articulated across various social contexts. This will require distinctly different transcription of the interviews or focus group discussion than research on speech patterns.
One then must decide what you needs to be transcribed,especially if you have collected qualitative data.Choices concerning whether a transcription should include nonlinguistic observations (facial expressions, body language, setting descriptions, etc.), whether transcription should be verbatim (it is recommended, by most transcription protocols, to produce at least one verbatim transcription of the entire interview done by a professional transcriber.), and whether to identify specific speech patterns, vernacular expressions, intonations, or emotions are to be made before transcription begins.
Capturing the contextual milieu.
Be aware that transcribing is an ongoing analytical process; textual data is not the “rock-bottom” of interview research but is an artificial construction that turns a speech presentation into written form. Equally essential to the primary textual data are contextual notes denoting and explaining pauses in speech or non-linguistic communication such as facial expressions. The relationship between audiotape and transcript is the relation of a “realist” object versus a “constructivist” object. One’s data is ultimately found at the intersection of these two objects.
Maintaining the naturalness of speech.
Practical difficulties in preserving the naturalness of speech include: elisions, incomplete sentences or overlapping speech and grammatical emphasis. The choices made must be based on the “analytical contribution” they make to the theoretical framework and specific research question(s). Many aspects of speech are difficult to capture through text such as speech elisions, incomplete sentences, overlapping speech and lack of clear-cut endings in speech. Moreover, close attention must be paid to how punctuation is inserted in order to avoid adding emphasis where there was none in the original speech presentation. Researchers must be very clear with transcribers about what they want included.
Whatever the level of analysis applied, all transcription processes will benefit from an organized and systematized storage protocol. Research “pandemonium” can be avoided by maintaining backups, ordering field notes accordingly, designing and implementing a labeling and logging system and cataloging items. More information on managing your research data in the next post.
Source: McLellan, Eleanor, MacQueen, Kathleen M., and Neidig, Judith L. 2003. “Beyond the Qualitative Interview: Data Preparation and Transcription.” Field Methods 15(1): 63-84.