Applied Linguistics at KMUTT

Here's a methodological question for all of you: this is something that has come up in almost all my articles.

If you are quoting a participant, how close should one be to the original quotation. With oral data, there are a lot of false starts and disfluencies - should these be reported or 'cleaned up'. This is perhaps more important when the participant is a non-native English speaking teacher of English. By staying close to the original quote, you stand the risk of making them look incompetent as English speakers, possibly. By 'cleaning up' what they say, and making it more grammatical or coherent, however, do you run the risk of losing some of the authenticity of the data, and more importantly, of imposing your meaning onto what might otherwise possibly have meant something different.

As an example, here is faithful rendering of an interview with a director of studies answering the question of what content he focuses on during an observation.

"Okay look for, first one, look for okay the teachers manage the classroom and the teaching approach the teachers use okay whether student centred or teacher centred approach. Okay and also we check whether the teachers are prepared before the lesson or not okay "

Even the full stop/period there is questionable, as this was pretty much a run-on stream of words - I thought, perhaps, that I may have just caught the trace of an intake of air at that point rather than just a pause, indicated by the commas elsewhere.

Should this be presented as is?

Can we take the easy way out of answering this question and say that "it depends on what you are trying to show"?

What is the purpose of including this quote in your article? If you just want to share the content of the quote, then it seems to me that you should be able to edit it just for readability's sake. For example, the speaker doesn't start answering the question until the second "look for", so everything before that can be deleted. And it seems like "Okay" often means full stop, comma, or a missing word in this person's speech.

So, would it be ok to write this as:

Look for [how] the teachers manage the classroom and the teaching approach the teachers use, whether student centred or teacher centred approach. And also we check whether the teachers are prepared before the lesson or not.

Basically, I'd argue that if you only care about the content of what is in the quote, then readability is the priority. I don't think this edited example loses any authenticity or meaning. But this is not my field at all, so I'm really just guessing. It's an interesting question so I'd love to hear from others who have interview data in their thesis.

Stuart

I was thinking about this right after the most recent progress presentation we had.

In one feedback I received on my paper, a reviewer asked why I included all the false starts, discourse markers, and grammatical disfluencies even when these features did not affect how I looked at my data. I am assuming that the reviewer asked me this was based on the belief that how the data was represented should reflect the analytical approach.

When I look at the core papers that I used for my study (Trent, 2012; Luk, 2012), their data (excerpts of interview data) looked very 'neat' (and correct). Nonetheless, the level of correctness is doubtful because my interviewees who were themselves native speakers of English were not 100% fluent and accurate in their language use during the interview sessions.

I would say it depends on what you want to show in your data? (Just because the reviewer had brought this up when I submitted my paper).

I actually had similar feedback from one reviewer who asked if my respondent was a stutterer. He wasn't, but on paper it certainly looked like he could have been. I guess that being faithful to the transcription doesn't work since you can't transcribe the speed of delivery, or the way in which interruptions might be the cause of the disfluencies in the first place.

In the end, I think Stuart is right. It depends. The clarity of the content of the quotation is most likely to be important unless there are specific elements of the disfluency that is worth highlighting for the argument of the paper. In addition, on a more practical level, including all the false starts impacts word count, and when you are under pressure to keep your particle under a specific length, those extra words probably have more value elsewhere.

Assuming that the "it depends" of Stuart concerns a focus on the content (rather than the delivery), it does seem that giving a detailed transcription including false starts, hesitations etc. would detract from conveying the content - the goal of the quote, yet it can be difficult to decide how much to 'simplify' the transcription. An alternative approach that should keep everyone happy is to give the 'simplified' content-oriented quotes in the article itself (keeping the word count down as Steve says) with a couple of sentences in the methodology explaining that the quotes are 'simplified'. To allow a reader to make their own judgment, the full complex transcript could be given in online supplementary materials.