Transcription notes
Transcription Script
The observers used conventional orthography to transcribe what they heard. There is a high level of consistency in how they did this, but some individual variation. In the corpus files I have kept the transcription as it was in the original, apart from correcting very obvious mis-spellings. There is a glossary attached below (transcription system) showing the transcriptions used to capture the local dialect and accent and to render weak forms.
‘Cleaning’ the data
The only way I have doctored the data is to add headings and to add the observer’s question(s) in the case of interview data. In the original data, the question is often only stated in the first interview and not repeated for subsequent exchanges. Where the question can easily be inferred (and observers were encouraged to ask questions in exactly the same way), I have simply added it. Where reconstructing the original form of the
question needs a high degree of inference, I have included it between < >
Coding of the data
I have used the following simple coding (mostly as in the original) to record the metadata between < >
Gender: F = Female M = Male
Social Class: A, B, C, D = social class as represented in the original: A = upper class B = professional; C = skilled working class; D = unskilled working class
Age: The observers generally only gave ages to the nearest 5 years e.g.< F50D>; <M35C>
‘Stage directions’ are also included between < > e.g. <laughs>
Speakers: To indicate different speakers in the same dialogue, I have numbered M and F speakers separately, as below. Where there was no information about the speaker, I have used <Sx>. Where there are isolated comments which do not seem to be part of a wider conversation, I have used <IC> for isolated comment
<F1> It’s very nice isn’t it <looked at wardrobe.>
<F2> N: Carpet’s nice isn’t it.
<F1>: Eh, it’s beautiful.
<M1> No so bad at I/II a yard.
<F1> They didn’t go to Woolworth’s for that! <Laughs>
Photo©Bolton Council
