CMB Qualitative Inquiry

ix sagepub com Qualitative Inquiry ix sagepub comcontent112226 The online version of this article can be found at DOI 10 11771077804273413 2005 11 226Qualitative Inquiry Cindy.ix sagepub com Qualitative Inquiry ix sagepub comcontent112226 The online version of this article can be found at DOI 10 11771077804273413 2005 11 226Qualitative Inquiry Cindy.

Learned to Love Transcription

Cindy M Bird

State University of New York–Fredonia

As a novice qualitative researcher, the author had little idea of the significant and vital role of transcription in the qualitative research process until she undertook her own tran- scription tasks The questions she asked herself during that work led her to insights on transcription both as product and as methodological process Her purpose in writing this article is to enter into the discussion of the nature and role of transcription in qualitative research by contributing an experiential context created when the transcription process

is regarded as a key phase of data analysis within interpretive qualitative methodology The structure of this article parallels the sequence of her learning experiences The find- ings generate an enhanced view of transcription and transcriber as methodological com- ponents, provide a clearer vision for researcher positioning on the intersecting land- scapes of transcription and qualitative methodology, and illuminate some implications for the training of qualitative researchers.

Keywords: transcription; data analysis; qualitative research;

narrative/autobio-graphical; research lens

With several audiotapes and piles of printed transcripts in front of me, I sat

back and sighed, “Wow, that was a learning experience.” The knowledge

cre-ated from that learning experience was not just the pragmatic type of ing a new skill; it was also the type gained through reflective contemplation ofthe learning process itself As a novice qualitative researcher, I had little idea

acquir-of the significant and vital role acquir-of transcription in the qualitative research cess Not until I undertook my own transcription did I begin to realize themagnitude and significance of this activity, both in its pragmatic and theoreti-cal states Subsequent reading of literature about transcription brought me

pro-Author’s Note: I wish to acknowledge and say thank you to the research participants who allowed me to use a part of their transcription in this article I would also like to thank those experienced transcribers and qualitative researchers who commented on the drafts of this article, those reviewers who offered very helpful suggestions, and the professor who pushed me into this experience.

Qualitative Inquiry, Volume 11 Number 2, 2005 226-248

DOI: 10.1177/1077800404273413

© 2005 Sage Publications

further awareness and insights into transcription issues and theircomplexities.

My purpose in writing this article is to enter into the discussion of thenature and role of transcription in qualitative research My contribution is

to provide both an experiential context and a knowledge base created whenthe transcription process is regarded as a key phase of data analysis withininterpretive qualitative methodology The context for this learning is myexperience of personally transcribing three different types of recorded data: aone-on-one interview, a small group discussion, and a classroom lesson Thisarticle begins with a view of the landscape of transcription and then examines

my experiences, questions asked, decisions made, and knowledge gained Itpresents a perspective on transcription attained through personal experi-ence and a positioning within the even larger landscape of qualitativemethodology

THE LANDSCAPE OF TRANSCRIPTION

Transcription as Data

Fundamental to the concept of transcription is its ontology How ers perceive of transcription affects their treatment of it within their researchmethodology Bogdan and Biklen (1998) thought of transcription as a datacollection method, and the resulting transcripts as raw data, on a par withparticipant observation field notes Green, Franquiz, and Dixon (1997) alsoheld the idea of transcripts as data but focused on their constructed quality:

research-A transcript is a text that “re”-presents an event; it is not the event itself ing this logic, what is re-presented is data constructed by a researcher for a par- ticular purpose, not just talk written down (p 172)

Follow-That “particular purpose” hinges on the methodological location of theresearcher

Transcription as Act

Most qualitative researchers will agree that transcription is the act of(re)presenting original oral language in written form But what type of act isit? To Roberts (1997), the act is political Just as talk is a social act, so too is tran-scription In trying to “fix the fleeting moments of words” into written form,transcribers are evoking “the social roles and relations constituted in lan-guage” (Roberts, 1997, p 168)—language ideology as they conceive it Atran-

scriber constitutes a social and political being; any act of transcription duced by such a being must of consequence be subjective.

pro-With political concerns also come ethical concerns Viewing transcripts asnonobjective constructs, ethics asks how the voices of the research partici-pants can be heard in the way they wish them to be heard When “every deci-sion about how to transcribe tells a story,” the question is, “Whose story?”(Roberts, 1997, p 169) Roberts (1997) asked this ethical question from hermethodological place of linguistics However, such a question also holds akey place in the methodology of narrative and other interpretive inquirywherein the focus is less on the language employed and more on the intentand context of the language Clandinin and Connelly (1994) emphasized theneed for all stories to be recognized and acknowledged and all voices to beheard Here, “voice” is more than verbal sound and authentic dialect; itincludes social context and embedded and intended meaning Thoseresearchers who desire to present the voice of the participant must wrestlewith this issue while they knowingly write the research as their own product.When representing an oral voice in written form, the transcriber becomes the

channel for that voice Because the transcriber is not that voice, any act of

tran-scription becomes an interpretive act, as Lapadat and Lindsay (1999)proposed

Transcription as Interpretive Act

The choices that researchers make about transcription enact the theories they hold and constrain the interpretations they can draw from their data (Lapadat & Lindsay, 1999, Introduction section, para 1)

tenHave (1997) strongly emphasized the idea that because transcriptionscannot represent all details of a recording, they are “always and necessarilyselective” (Interest section, para 7) This act of choice implies decisions aboutsignificance, which in turn imply interpretation from some point of view.Green et al (1997) further explained that transcription is not only an interpre-tive act but also a situated act As researchers locate themselves within thecontext of their own assumptions about language and culture and discoursepractices, they are more able to recognize their own interpretive acts, tounderstand that “writing down what one hears is the result of a range of inter-pretive acts” (Green et al., 1997, p 173) To borrow an example,

Hearing a sound as stress involves understanding how stress is signaled and understood within a particular language group To see silence as meaningful, and not merely the absence of talk, or to see someone as taking the role of ques- tioner involves understanding of the discourse practices of a social group (Green et al., 1997, p 173)

Although many of these interpretive acts flow from the social, cultural, andlinguistic location of an individual transcriber, others are grounded in thetranscriber’s methodological stance (Note that in this section, I am using

transcriber and researcher interchangeably Later I will speak of the situation

where the transcriber is not the researcher.)

Some qualitative researchers seek to give a holistic picture of the event oroccasion they are studying They place great emphasis on the everyday expe-riences of individuals to create “pictures” or “portrayals” (Fraenkel & Wallen,

2000, p 12) Other qualitative researchers focus more on contextual tion of interest to their methodological stance These researchers from a morephenomenological perspective seek to focus both on transcription as an inter-pretive act and on transcriber as interpreter However, researchers within thedomains of conversation analysis (CA), discourse analysis (DA), and the lin-guistic methodologies tend to focus exclusively on conversation and havebeen criticized for neglecting contextual factors of the conversation event(Lapadat & Lindsay, 1999; see tenHave, 1997, for a defense of the contextualinterests of CA/DA researcher-transcribers)

informa-Transcription as CA and DA

Researchers in the field of CA take as their goal the description and cation of socially organized interaction, the “ordinary chit-chat of everydaylife” (tenHave, 1997, Methodological section, para 7) The transcription sys-tem used in CA is “specifically designed to reveal the sequential features oftalk” (tenHave, 1997, Interest section, para 7) Closely associated with CA is

expli-DA, wherein transcription is “the process of creating a written representation

of a speech event” (DuBois, Schuetze-Coburn, Cumming, & Paolino, 1993, p.45) Indeed, their association is so close that transcription conventions mayapply to both (Edwards & Lampert, 1993)

Transcribers working within these analytic methodologies have available

to them conventions and notations that have been developed for and withinthese fields Psathas and Anderson (1990) outlined some practices of tran-scription, and tenHave (1997) provided an appendix of transcription notationspecific to CA, such as using punctuation marks to indicate intonation: aperiod indicates a stopping fall in tone, a comma indicates continuing intona-tion, a question mark indicates a rising inflection, and a dash indicates anabrupt cutoff Perhaps foremost, DuBois et al (1993, pp 55-65) provided tran-scribers with a plethora of conventions and symbols for the transcriptionproduct, including speaker tone [/ rising, \ falling]; pause [ for break inspeech rhythm, for brief pause of half second]; vocal noises [(SWALLOW)];and transcriber comments [((very adaptable I believe to most methodologicalperspectives))]

Although CA and DA share conventions unique to linguistic ogy, this does not mean that their transcribing conventions are so project spe-cific as to be of little value to transcribers in other qualitative methodologies.Edwards (1993) noted that from the many taxonomies of transcript proper-ties, two general design principles can be recognized:

methodol-1 that the transcript preserve the information needed by the researcher in a ner which is true to the nature of the interaction itself

man-2 that its conventions be practical with respect to the way in which the data are to

be managed and analyzed (p 4)

In principle number one, the “information needed by the researcher”depends on the researcher’s methodological approach, whether linguistic,conversational, or phenomenological Thus, it would seem that the work ofthe transcriber (whether researcher or hired contractor) must also be deter-mined by and dependent on the researcher’s methodology It would also fol-low, then, that my chosen methodology, being less linguistic and more quali-tatively interpretive, would determine the interpretive and analytic nature ofthe work of the transcriber

Transcription as Interpretive Analysis

“Interpretation,” here, refers to the effort to formulate the relatively unique meaning an utterance, an action or an episode seems to have for participants and/or researchers, while “analysis” is used to indicate efforts to isolate aspects, mechanisms and procedures that are relevant to a range of cases (tenHave, 1997, Discussion section, para 1)

Although much literature exists about transcription methodology withinthe more language-focused research fields (see, e.g., tenHave’s Web site athttp://www.pscw.uva.nl/emca/index.htm), little has been presented on thenature of transcription in a wider qualitative methodology One influentialexception is the work of Lapadat and Lindsay (1999), which aims for a morecomprehensive approach to transcription: from the practice and its standard-ization of techniques to the theoretical and its interpretive positioningswithin a range of qualitative methodologies

In agreement with tenHave’s (1997) definitions, Lapadat and Lindsay(1999) maintained that the act of transcription is interpretive in that it is theprocess of formulating and producing a meaning unique to the situation ofutterance As such, transcription “facilitates the close attention and the inter-pretive thinking that is needed to make sense of the data” (Lapadat &Lindsay, 1999, Conclusion section, para 6) However, they also maintainedthat analysis begins during transcription, thus, making transcription “anintegral process” in qualitative analysis For this analysis to meet the

demands for rigor in qualitative research, they agreed with Poland (1995) incalling for “greater reflectivity” (Lapadat & Lindsay, 1999, Contextualizedsection, final para.) on the part of researchers regarding the transcription pro-cess As an example of such reflectivity, they said that researchers should askthemselves this question: “What is a useful transcription for my research pur-poses?” (Kvale as cited in Lapadat & Lindsay, 1999, Contextualized section,para 7).

Transcription as Product

Edwards’s (1993) second principle calls for conventions to be “practical”

as they relate “to the way in which the data are to be managed and analyzed”(p 4) In line with this principle of practicality, Psathas and Anderson (1990)set a goal of establishing a standard set of conventions to enable comparisonsacross CA studies (Lapadat & Lindsay, 1999) A few years later, tenHave(1997) compiled a group of notation conventions designed to relate tran-scripts to the data analysis methods of CA He explained his symbols “insummary fashion” (appendix) from the work of Atkinson and Heritage(1984) who in turn worked from the transcription notation system developed

by Gail Jefferson (Atkinson & Heritage, 1984, p ix) The Jeffersonian scription System developed in the early 1970s as Jefferson began the task oftranscribing a series of taped academic lectures (Jefferson, 1992) She created aseries of notations designed to capture the social situation of a lecture hall,including student questions and professorial comments unrelated to the lec-ture topic per se, such as “What time is it?” Later, she edited these transcripts

Tran-to publish them in book form (Jefferson, 1992)

Of note is the fact that although the above researchers were working marily in the fields of CA or DA, they were unable to determine a definitiveset of transcription conventions for those fields, let alone a set applicable tothe larger field of qualitative methodology Theirs and other adaptationsthrow into question the idea of a “universal” set of transcription conventionsfor qualitative research As convenient as it may at first seem, however, hav-ing such a standardized set “is not theoretically tenable” (Lapadat & Lindsay,

pri-1999, Search section, para 2) given the flexibility required by researchers toapproach transcription in ways suited to their different methodological per-spectives Instead, researchers who transcribe must determine their own con-ventions for the product of their transcription Because the methodologicaltheories they hold will determine the answers to the transcription questionsthey ask, perhaps it is the questions that carry the more universal nature

As noted earlier, I have been using transcriber and researcher

interchange-ably in this section: the assumption being that the researcher either performsthe actual transcribing work herself or himself or has direct supervision of thetranscribing, even to the point of providing a list of preferred conventions

However, not all transcribers are researchers, nor are they given some form ofdirect supervision The section below examines this issue of a transcriber who

is someone other than the researcher

Transcription as Peripheral Participation

My experience with transcription to date has been as a transcriber As well, I had been present at the taping of all the communicativeevents I was required to transcribe However, that type of positioning is notalways the reality in research It is common for someone other than theresearcher to do the transcribing (Lapadat & Lindsay, 1999; Tilley, 2003a).Many times, especially on large research projects, the transcription is con-tracted out to a transcriber who is not part of the research team and who wasnot present at the recording Consequences of this type of transcribing eventare explored in the work by Tilley (2003a, 2003b)

researcher-Using the concept of “legitimate peripheral participation” (Lave &Wenger as cited in Tilley, 2003b, A Situated section, para 1) as an analyticalframework, Tilley (2003a, 2003b) explored the complexities of the transcrib-ing event when researcher is not also transcriber Her findings reveal thepotential for educational experiences for those involved, both thenonresearching transcriber and the nontranscribing researcher Tilley’snonresearching transcriber was given a few general guidelines; however, theinterpretive nature of transcription forced the transcriber to make decisions

as the transcribing process progressed Because, as Tilley (2003a) argued, “Atranscriber’s interpretive, analytical, theoretical prints become visible uponclose examination of the transcription process and the texts constructed”(p 752), the nontranscribing researcher who studies these transcripts anddiscusses them with the transcriber has a greater opportunity to gain “a morecomprehensive understanding” (Tilley, 2003a, p 771) of and an added dimen-sion to the data analysis and interpretation phase of the entire researchproject

LOCATION ON THE LANDSCAPE

My literature search yielded two results First, my knowledge about scription increased I learned of conventions and issues of product creation,but I also moved from a purely technical consideration of transcription into aconsideration of its theoretical landscape Second, as the landscape of tran-scription took shape before me, I found myself becoming more aware of myown position on that landscape In the field of qualitative methodology, Icould already locate myself near the interpretive and phenomenological area.Now with a view of the transcription landscape as well, I can recognize an

tran-intersection of the two and wave hello to my CA neighbors from myinterpretist plot of ground.

What my search did not reveal, and what I had yet to experience, was theissue not of location but of equipment, the machines that would assist mysojourn in the land of transcription To transcribe my first cassette tape, I used

a tabletop-sized, dual speaker, single-cassette player with a pause button but

no counter Not until later in the story of my transcription experience did Ilearn what a transcription machine was and how to use it Thus, in the begin-ning, expending well more than 40 hours to transcribe a 1-hour tape did much

to build my dread of transcription

After I had complained much to my fellow researchers, one of them finallytold me about transcription machines, how they worked, and where I couldborrow one So by the time of my second transcription event, I had obtainedthe use of an actual transcription machine This older, somewhat bulkymachine came replete with oversized foot pedal and cracked earphones.However, once I learned to adjust the controls to my liking, and operate thefoot pedal in my stocking feet, my transcribing life was forever changed—theoverall time for transcribing became significantly reduced More to myresearcher liking, however, was the ability to play the same few seconds oftape over and over again until what I was hearing made some sense syntacti-cally and held some contextualized meaning for me Although Poland (1995)suggested that this process of repetition creates unintentional mistakes, Iknew that my strong assumptions about language and meaning would push

me to try and retry until I found some meaning contextualized in what Ithought I heard Thus, the transcription machine, with its special features forlistening to recorded tapes, helped to ease some aspects of transcription; theprocess was now becoming less dreaded Perhaps I would learn to lovetranscription after all

THE NARRATIVE OF MY TRANSCRIBING

Clandinin and Connelly (1994) contended that narrative “names the tured quality of experience to be studied” (p 416) The following narrative of

struc-my transcribing experience is indeed structured, for I do not want to bore thereader with tales of tired ears from wearing headphones, or of sore back andleg muscles, or of the long hours of rewind button pushing before I acquiredand gained a degree of competency with a transcription machine

Instead, I weave my remembrance of three separate transcribing eventsinto one “narrative of experience” (Clandinin & Connelly, 1994, p 416) Ineach of these events, I was researcher, participant observer, and transcriber.Each transcription experience contributed not only to my own experience oflearning to transcribe but also to the developing and honing of my awareness

of transcription as a key element in data analysis The construction of this

nar-rative further developed and honed my reflective thinking both on the scription process and the reflection process.

tran-The One-on-One Interview

I clipped the lapel mike on the neck of his T-shirt as my participant sat inthe lawn chair on the veranda of his house overlooking the sparkling water ofthe lake on that warm July day I had come to ask this retired teacher somequestions about teacher professionalism and curriculum reform during thepast three decades The research was for a paper for a course on curriculumfoundations

The interview lasted about 1 hour The transcription process took muchlonger

I began my transcribing concerned with the content only, the words andphrases this teacher used to impart the information I sought Prior to theactual interview, I had e-mailed the teacher a set of questions that I would ask

He had written notes in answer to those questions and preferred to followthat order rather than have me ask further questions So I started typing mytranscript remaining faithful to his carefully chosen words—his thoughts,ideas, and experiences expressed in linguistic form

Not long into the transcribing process, I encountered my first tion issue”—the paradox that this teacher’s words conveyed one meaningwhereas the tone of those words conveyed another I had not considered thisdifference while interviewing because at that time I had had an understand-ing of his meaning, so I thought little more about his tone As I listened to thetaped interview, however, his tone seemed to become more evident Perhaps

“transcrip-it was his tone plus my memory of the event E“transcrip-ither way, I found myself lyzing the data and reflecting on his tone and its significance to the overallmeaning of his lived experience His phrase “that was fun” sounded like “that[emphasis] was fun.” The tone, as I understood it based on my participation

ana-in the situational context of that time ana-in the ana-interview, was one of sarcasm,intended to convey a meaning opposite to that of the words themselves Thequestions then became, Do I ignore tone? Do I record tone?

To make the decisions, I returned to my original intention for interviewing.The topic was curriculum reform, and my goal was to obtain the teacher’sperspective By interviewing a retired teacher who had lived through andexperienced several reform initiatives in the province of Ontario, Canada, Ihad hoped to capture a sense of how it was for him as he had experienced it

The title of my paper was to be Curriculum Issues of Today as Outgrowth of the

Past: Lived Experiences as Context for Curriculum Change My thought was that

to attain an “as lived” perspective, I would need to record more than his fully generated list of initiatives and how he had seen them play out in prac-tice in the schools I would need to record the affective nature of his experi-

care-ence as well I would need to analyze more than just the words of thatconversation So the tone stayed.

Near this point, I began to realize that transcription does indeed includeelements of analysis and interpretation, as Lapadat and Lindsay (1999) main-tained In this case, where the words were written beforehand, what was thepurpose of interviewing if the only interest was the words? What was thevalue of face-to-face communication if not to add my own observations andinterpretations of a communicative event? By asking these questions, I hadbegun the reflective process: creating “greater reflectivity” on the transcrip-tion process itself and adding rigor to my qualitative research (Lapadat &Lindsay, 1999) By turning to my original intention for interviewing, I had,without conscious realization, followed Lapadat and Lindsay’s example ofreflectivity by the researcher-transcriber: asking the question of what consti-tutes a “useful transcription” for my research purposes (Contextualizedsection, para 7)

Once the decision about tone was made, a subsequent, practical issueemerged How would I write tone? Not having read any material on conven-tions for transcribing, I used my knowledge of conventions for quoting liter-ary material Because I would be inserting an “editorial” comment into theoriginal material, I chose to use square brackets around the comments Theresult was “that [emphasis] was fun [sarcastic tone].”

As I returned recently to my original transcript of this interview, I saw anumber of invented conventions that revealed an analysis in progress—or asTilley (2003a) called them, the “transcriber’s interpretive, analytical, theoreti-cal [finger]prints” (Introduction section, para 4) I had made a number of ana-lytical choices for the actual typing of the transcription, choices that material-ized into my conventions First, from my desire to get the information andstick to the topic, I omitted those phrases and sections of the original inter-view that I deemed as “nonrelevant.” I inserted an ellipsis for such omissions.For times where the interviewee strolled down a tangent, I followed the ellip-sis with a summary note in square brackets Also knowing that grammati-cally speaking an ellipsis could show a hesitation or pause in speech (Harris,

1997, p 184), I decided that differentiating between the two in the transcriptwould be of timesaving value by preventing the need to return to the tapeitself to ascertain whether the meaning was omission or pause Therefore, I

chose to use the ellipse only for omission and create the convention of [pause]

to indicate a break in speech

The following examples indicate my conventions for analytic choices:

1 Statistics, algebra, you take, you have no reason to use it, and on and on [pause]

2 And it’s sad because [story of young teacher who lasted only 2 years in teaching]

3 It’s a lot of hot air [pause] It really is [chuckle].

4 the purpose of being there, and that was to work with kids [pause]

5 And teachers wanted to get on with [pause], well, they wanted to know what had to be

taught and when it had to be taught.

As well as adding summary notes inside square brackets, I also added rial comments Most often the comments were points of clarification:

edito-1 Now in the early stages [of my career] I didn’t see any great curriculum issues.

2 [he came from industry and thought] what a disorganized shambles this [education] is.

3 [summary of minimum knowledge] it’s teaching kids how to learn How to read properly,

to summarize the stuff they’ve read, and thus accumulate [their own] body of knowledge.

However, there was the occasional grammatical correction:

So you knew EXACTLY what you had to do THEN with Hall-Dennis, a total[ly new] ball of wax came in.

(Note: As I inserted this section from the original transcript, I found myselfasking why I had felt it necessary to make that grammatical correction—further evidence of the multiple levels of analysis involved in reflection andnarrative inquiry I also found myself chuckling at the irony of the situation.Here was a transcriber who was most willing to omit chunks of oral data from

a transcript because she felt the content was not “relevant,” yet she felt pelled, without a second thought, to record the relevant data in correct gram-matical form Perhaps I have some print-structure assumptions to go with mylanguage-meaning assumptions.)

com-Another of my typing conventions was using capital letters to indicateloud volume I had thought about using italics but dropped that thoughtwhen I remembered that italics and other such computer formatting are oftenlost when transferring files from one format to another:

1 And they wanted to spend time with kids doing the REALLY worthwhile things.

2 And how DO we decide what we are going to teach?

3 Imagine calling a math curriculum meeting DURING school hours.

Sometimes, the recorded words were just too indistinguishable or inaudible,

or too many people talked at once—as when his wife shouted an answer frombehind the screen door in the kitchen Rather than guess a word or leave abreak in the transcription, I decided to employ a convention indicating thatthe problem was with the data collection process and not the transcriptionprocess:

and the PA days and the here-we-go-again [unclear].

The conventions I had actually used as a result of decisions made came morefrom my personal experience with grammar, quotation formatting, and aca-demic writing and less from transcription itself