Methodological Considerations: Transcription as the Act of Representing, Analyzing, and Interpreting ‘Talking Data’

Voice Recorder

Handoyo Puji Widodo (Discipline of Linguistics, University of Adelaide)


Interview data transcription is part of the qualitative research activities designed to capture and unpack the complicatedness and meaning of naturally occurring phenomena (e.g., values, beliefs, feelings, thoughts, experiences) in social encounters. It becomes the norm in most qualitative research studies. Literally, transcription is a useful means for turning digitally recorded interview data (findings) into transcripts, but methodologically speaking, transcription is the act of representing original spoken text (recorded talking data) in written discourse as well as analyzing and interpreting instances of these data (Bird, 2005). These data in the form of transcripts are viewed as text, jointly created by research participants and a researcher through dialogic conversation and negotiated engagement. In other words, transcription is seen as the act of data representation, analysis, and interpretation, and indeed it is an activity that requires sound methodological orientation. In response to this, I would like to briefly discuss some methodological considerations in data transcription to help emerging or beginning researchers prepare transcripts on the right track.

Listening to Talking Data

The first step in organizing and analyzing talking or verbal data is doing the transcribing, which involves close observation of data through carefully repeated and attentive listening.

Transcribing verbal data is a useful starting point for data organization and analysis at the outset for some reasons.

Firstly, transcribing verbal data affords a researcher the opportunity to carefully listen to, pay close attention to, and think deeply of digitally recorded data situated within a particular interview context. This socio-cognitive activity involves how researcher’s mind interacts with spoken text.

Secondly, as Anderson and Jack (as cited in Matheson, 2007) point out, transcribing own data “provides a unique opportunity for [researchers] to critique their own work and potentially improve upon their interviewing technique“ (p. 549).

Thirdly, listening to verbal data allows a researcher to analyze what emerging finding themes should further be examined and enables her or him to reflect on what she or he has asked to the participants. In other words, listening attentively to verbal data allows researchers to get closer and more familiar with the data.

Shaping Talking Data

How researchers shape or present verbal data in written form depends, to some extent, on transcript layout choices.

I would like to suggest some layout features of transcripts.

To begin with, researchers should provide data identity (e.g., data code & number, data collection date, involved participants, data collection methods). This information enables researchers to easily retrieve the data and allows for tidily organized data management. As exposure to participant’s identity is much ethically concerned, a researcher is advised to assign a pseudonym to the participant’s name. This ethical issue needs to be spelled out before interview sessions commence.

Secondly, researchers had better leave space for transcription symbols to give the reader with “a set of [spoken] conventions for displaying actions and utterances in naturalistic situations” (Ochs, 1979, p. 61). These symbols help the reader read talking data.

Thirdly, a researcher should structure transcripts into line-by-line dialog to show turn-by-turn dialogic interaction between participants and herself or himself.

Further, the names of participants need to be coded by one or two letters possibly along with one or two numbers to allow more space for dialog. What follows, researchers had better number each line-by-line dialog to make data analysis easier. If participants and an appointed critical debriefer (peer) wish to verify information taken from a transcript, they will easily double-check the information in the original transcript. Moreover, a researcher has to leave space for data feedback, verification, and accuracy when member checking is done. Thus, the features of a transcript layout assist researchers to better analyze particular information and re-examine this information for further emerging finding theme examination.

Communicating Talking Data with an Interpretive Intent

Communicating talking data means detailing and interpreting them in a methodologically sound manner. This involves how much detail talking data should be transcribed. As a rule, this transcription practice involves two levels or approaches of transcription: naturalism and denaturalism. The former suggests that every instance of utterances should be transcribed in greater detail, but in the case of the latter, idiosyncratic elements of speech (e.g., stutters, pauses, fillers, non-verbal signals) need to be removed.

These two methodological positions correspond to certain views about the representation of a language. In a naturalized approach, a language construes and represents real-life phenomena (Schegloff, 1997), but what matters in denaturalized transcripts is that meaning and perception construct one’s reality.

The choice of a transcription approach, either naturalism or denaturalism, depends on a degree of reflexivity required, data analysis focus, and sensitivity to participants and the nature of their involvement in the research.

However, researchers who are willing to look closely at different aspects of talking data, a naturalized approach would be a good fit because this detailed transcription shows the complexity of the transcription process, maintain representation or authenticity of lived experiences, and modulate the interpretation of transcription data at a given delicacy level (Tilley, 2003).

It is important to bear in mind that in talking data transcription, any grammatical mistakes are left uncorrected because spoken discourse may comprise grammatical mistakes, which sound natural as long as the intended meaning is successfully communicated or understood. In this respect, this transcription goes beyond the norm of standardization regarding language accuracy.

Regardless of the transcription approaches, researchers should be responsible for making meaning of or unpacking verbal data with an interpretive intent. This interpretative intent facilitates them in (re)constructing verbal data in relevant depth and breadth.

Reproducing or (Re)constructing Talking Data

Due to the advent of recording technology, audio and video recordings become a common practice of capturing or collecting verbal data in qualitative research. Digital recordings ease researcher’s work regarding data transcription.

Transcribing data does not simply mean (re)producing accurate transcripts, but communicating inner voices of research participants naturally and credibly (Hammersley, 2010).

Transcribing verbal data is not sort of data (re)production because this implies that researchers exploit data from the exploited (research participants). Secondly, doing the transcribing as data (re)production suggests that researchers control how data are collected during interviewing. In an ethical sense, I argue that doing transcription is a way to (re)construct talking data because participants and a researcher jointly create data through socio-cognitive encounters (dialogic conversation or collaborative dialog). In this sense, participants are seen as co-actors who have a legitimate ownership of knowledge and help a researcher better understand realities or inner voices of the participants. Drawing on this notion, both participants and a researcher play a pivotal role as data co-constructors whose responsibility is to unpack the multiplicity of knowledge, voice, and experience from different angles. For this reason, before fieldwork commences, it is useful for a researcher to spell out levels of involvement or roles both participants and she or he have to play to establish trust and show mutual respect.

Building Data Credibility

Credibility is a unique feature of empirical qualitative research. One of the main ways to achieve credibility is by performing a member check.

Member checking allows research participants to provide feedback on the accuracy of how talking data are presented and interpreted. Buchbinder (2011) suggests another term ‘validation interview’ as a means of facilitating “a dialogue between [research participants] and [a researcher] intended to confirm, substantiate, verify or correct researcher[’s] findings” (p. 107). I contend that whether a researcher uses the term, either ‘a member check’ or ‘validation interview,’ relies upon a particular methodological stance. The most salient thing is that a researcher should be willing to communicate transcripts with research participants in order to achieve data credibility because research participants serve as the source of knowledge; for this reason, the data co-constructed with the participants should be communicated with them at debriefing meetings during and after fieldwork. Moreover, a researcher can appoint a critical debriefer who is well versed in research methodology and nature of research to look at whether the data are appropriately presented, analyzed, and interpreted. This etic or outsider’s input helps enhance data credibility.


Transcription is a powerful act of data representation, analysis, and interpretation in such a way that it exerts considerable influence on how data are conceptualized. Transcripts as a result of transcription should be well-organized, analyzed, and interpreted in order to represent and construe a set of social actions. In other words, transcription is a social and process oriented activity, which requires critical, reflective, and attentive thoughts at macro and micro levels of transcription. Indeed, it has implications for making an informed decision on organizing, analyzing, and interpreting research data (Lapadat & Lindsay, 1999).



Bird, C. M. (2005). How I stopped dreading and learned to love. Qualitative Inquiry, 11, 226-248.

Buchbinder, E. (2011). Beyond checking: Experiences of the validation interview. Qualitative Social Work, 10, 106-122.

Hammersley, M. (2010). Reproducing or constructing? Some questions about transcription in social research. Qualitative Research, 10, 553–569.

Lapadat, J. C., & Lindsay, A. C. (1999). Transcription in research and practice:  From standardization of technique to interpretive positionings. Qualitative Inquiry, 5, 64-86.

Matheson, J. L. (2007). The voice transcription technique:  Use of voice recognition software to transcribe digital interview data in qualitative research. The Qualitative Report, 12, 547-560.

Ochs, E. (1979). Transcription as theory. In E. Ochs & B. S. Schieffelin (Eds.). Developemental pragmatics (pp. 43-72). New York: Academic Press.

Tilley, S. A. (2003). Transcription work: Learning through coparticipation in research practices. International Journal of Qualitative Studies in Education, 16, 835-851.

Author’s bio:

Handoyo Puji Widodo, currently working on a PhD in language materials development at the University of Adelaide, has published edited books and presented his work at international conferences. His refereed articles and book reviews have also appeared in refereed international journals. He is currently sitting on an editorial board member of numerous refereed international journals (e.g., The International Journal of Innovation in English Language Teaching & ResearchAsian ESP Journal, TESL-EJ, The Canadian Journal of Applied Linguistics). Widodo’s areas of specialization include language curriculum and materials development as well as language teaching methodology. Email:


1 Comment

Filed under Academic writing, Methodology, Survey

One response to “Methodological Considerations: Transcription as the Act of Representing, Analyzing, and Interpreting ‘Talking Data’

  1. Nice information about qualitative research

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s