Benedict, *please* don't transcribe all those tapes!! NUD*IST doesn't know
or care how much you transcribe, and it certainly doesn't stop you thinking
about data that isn't in docfiles. I'm convinced one of the most serious
effects of qualitative computing is it somehow gave the message that in
order to reflect on data you had to get it into the computer. That's
basically because it gave the other (equally seriously wrong) message that
the only way to think qualitatively is to code. One of our problems early on
was that if you didn't transcribe everything you might get stuck later -
you'd coded the document and couldn't edit or insert without wrecking the
coding. But that doesn't apply now to NUD*IST4 (or of course to its big
sister, NVivo). (I assume you are using N4 since you said NUD*IST, but just
in case, here's the answer for both.)
What you transcribe is always (in any way, manual or computer, of handling
the data) a methodological decision - and it boils down to what do you want
to see again. Challenge in qualitative work is you can't be quite sure in
advance what will matter.
The data you want to have transcribed and imported into NUD*IST is the text
you want to see again in the new context of a category, theme or topic, or
the part of a document a search takes you to. The reason for transcribing is
that NUD*IST can then show you immediately, show you the document or
everything about the topic in a live screen browser, so you can review and
rethink everything on a topic, and code-on to finer topics, write memos
about the data using the quotes etc. It can take you back from any segment
to the context, to read and understand better. And it can give you the
material you want to analyse more finely, immediately, on the screen. That's
all precious - so we transcribe what may prove valuable when we see it
again. But that doesn't mean everything!
In any interview, there is lots of stuff you don't want to retrieve from
coding. Here's what to do with it.
1) If you want some of the interview transcribed, make a system of ways of
showing what's missing. If you don't want to see some parts ever (e.g. that
ten minutes when you talked about whether the tape recorder was working)
just summarize it in the transcript. But usually there is a possibility that
this might be useful later, and this mustn't be underestimated. So be
detailed in the summary, and always type the tape count number so you can
find the missing material fast.
If you find later that missing bit really mattered, you can insert it as a
new text unit and your coding will not be invalidated. (This is true of both
NUD*IST4 - N4 - and NVivo.) You can also append another transcription
section to the document in your project. So so long as you realise it
mattered, it can always be brought in later.
2) If you think you don't need this interview transcribed at all, be very,
very careful to record its existence in your project. Qualtiative research
is all about being surprised later by things you didn't think mattered.
Best way to do this is make an External Document in N4 (in NVivo a Proxy
Document) that records what the interview was, and what it offered and
where it is now and why you're not transcribing it. In NVivo this is a full
status, searchable, codable document so you can do a full summary and
partial transcript and hyperlink to the audio or video file. If you are in
N4 you can't make the compound doc this way, but you can put the description
into the header of an external document and write a memo about it that
describes it more fully, and then code the tape using the tape count number
as the text unit. (see pp 39-43 of the manual on the external doc and 63-65
for coding it).
Hope this helps
cheers
Lyn
Lyn Richards,
Director, Research Services, Qualitative Solutions and Research.
(email) [log in to unmask]
(Ph) +61 3 9459 1699 (Fax) +61 3 9459 0435
(snail) Box 171, La Trobe University PO, Vic 3083, Australia.
http://www.qsr.com.au
> -----Original Message-----
> From: Ki-Hoon Lee {PG} [SMTP:[log in to unmask]]
> Sent: Thursday, July 06, 2000 2:45 AM
> To: [log in to unmask]
> Subject: Is full transcription necessary?
>
> Hi, I am a new user of NUDIST and am encountering problems of huge
> transcription from tapes. I am not sure the efficiency and necessity of
> full
> description to code.
> I got about 90 interview tapes, and have quite challenging time to
> transcribe fully.
> Is it really necessary to fully transcribe or take major points (notes) to
> code and run NUDIST programme?
>
> Any help in this matter will be very gratefullty recieved.
>
> Thanks
> Benedict (K.H) Lee
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|