Reflection on transcription in early literacy contexts

Photo by arztsamui at FreeDigitalPhotos.net

Last night's hands-on work with Transana combined with my recent, background research about transcription have me wondering about my next big project with digital tools: using transcription software to prepare audio data from a Kindergarten literacy study.

  

Per my EP604 instructor's request, I have been doing some reading and reviewing of the literature on transcription, looking for methodological and theoretical insights that might inform my process. How should I approach this transcription job considering the context of the study?

The study in question is designed to assess young children’s incidental literacy learning as a result of exposure to and participation with moving picture books for children, aka "eBooks." Three examiners performed pre- and post-interviews with Kindergartners in which the students' interactions with traditional, print-based picture books were audio recorded. 

The design for this study is modeled after earlier studies performed by Sulzby (1985) and de Jong & Bus (2002, 2004).  So, I began my inquiry by taking a closer look at the methods sections within these articles. I am finding scant theoretical or procedural information provided about transcription, just as Lapadat & Lindsay (1999) asserted.

de Jong & Bus (2002) make no mention of how their transcripts were derived, saying only that they made use of "verbatim transcripts" during coding (p. 148).  However, it is clear from their description of data analysis that the transcripts included children's verbal and nonverbal cues for attending to text (for example, sounding out words as well as finger-point reading). In their 2004 study, the authors simply indicate that "verbatim transcripts" were analyzed (p. 387).  

In both studies, de Jong & Bus videotaped all book-reading sessions with the children.  I can only assume that the videos (and perhaps heavy fieldnotes) were tremendously helpful for fleshing out the transcripts. To me, this suggests broad implications and limitations for the current study in which I am involved, where no video data were collected due to significant IRB hassles.

Next, I checked out the 1985 article in which Sulzby describes findings from two studies that demonstrate a developmental pattern of children's emergent pre-reading skills.  She gives a fairly concrete description of her transcription procedure, even listing the symbols used to code the transcripts to indicate such things as finger-point reading, rise and fall of intonation, and phonetic spellings of attempted words.

Sulzby worked first from audio, then expanded the transcripts based on video and fieldnotes to include "descriptions of non-verbal behaviors coded to book pages and activities" (p. 476). According to Sulzby, the examiners who conducted and recorded the emergent readings were responsible for transcribing the tapes for each session. Each transcript was then checked by two other examiners.

That sounds like an ideal context in which to apply Transana software. Had Sulzby performed her study today, I imagine she and her research assistants would have found the multi-user version of Transana to be very helpful, given their commitment to the use of video and their collaborative process in which the transcripts were "continually checked during various analyses" (p. 476).

As I have no video to work from and very poor fieldnotes, I am continuing with my initial choice of Inqscribe software for purposes of transcribing the Kindergarten eBooks study.

However, I am curious about making more deliberate use of conventional transcription symbols (in addition to the slashes [ // ] I am already using to indicate the children's phonetic pronunciations).  One distinct advantage that Transana offers over Inqscribe is the integration of some basic symbol buttons in the text editor. I should make more concerted effort toward delineating pauses, changes in tone, and other vocal noises. According to Bloom (1993), "Studying language in context, and studying the development of language in the context of other developments in the child, require that we preserve far more than just the spoken word in the record we make of the data we collect" (p. 164).

For that reason alone, I wonder if Transana would not be an overall better choice for any project involving children's voices, regardless of the mode of data collection.

References (I am trying out the drag-and-drop function of Zotero and Mendeley!):

Bloom, L. (1993). Transcription and coding for child language research. In J. A. Edwards & M. D. Lampert (Eds.), Talking data: Transcription and coding in discourse research, 149-166. Hillsdale, New Jersey: Lawrence Erlbaum Associates, Inc.

De Jong, M. T., & Bus, A. G. (2002). Quality of book-reading matters for emergent readers: An experiment with the same book in a regular or electronic format. Journal of Educational Psychology, 94(1), 145. American Psychological Association. Retrieved from http://psycnet.apa.org/journals/edu/94/1/145/

De Jong, M. T., & Bus, A. G. (2004). The efficacy of electronic books in fostering kindergarten children’s emergent story understanding. Reading Research Quarterly, 39(4), 378–393. International Reading Association. Retrieved from http://www.reading.org/Library/Retrieve.cfm?D=10.1598/RRQ.39.4.2&F=RRQ-39-4-de_Jong.html

Lapadat, J. C., & Lindsey, A. C. (1999). Transcription in research and practice: From standardization of technique to interpretive positions. Qualitative Inquiry, 5(1), 64-86.

Sulzby, E. (1985). Children’s emergent reading of favorite storybooks: A developmental study. Reading Research Quarterly, 20(4), 458–481. International Reading Association. Retrieved from http://www.reading.org/Library/Retrieve.cfm?D=10.1598/RRQ.20.4.4&F=RRQ-20-4-Sulzby.html

Reflection on transcription in early literacy contexts