On Fri, 15 Dec 2006 09:53:45 +0100, Magnus Larsson <[log in to unmask]> wrote:
>I am not sure about what you mean with the possibility to code and
>analysis with these programs. At least with transcriber, the program I
>personally use, there are no such facilities as far as I've seen.....?
>
I think Transana has some coding functions but I'm not too familiar with
that program.
Transcriber does have lots of basic coding functionality. There's no
comparison with the QDA programs most people are familiar with but lots of
items are either automatically provided XML tags or can be manually tagged.
So speakers, speaker properties, speech turns and their times, etc. are all
given XML tags. There are also a host of tags for linguistic properties and
events (e.g. pause, laugh, noise). In more recent versions this has been
expanded to "named entities". The latter allows XML tags to be created for
different types of location, people, times/dates, organizations, etc.
There's a screen shot of this here:
http://trans.sourceforge.net/img/transcreen2.jpg. (Latter shot is from a
version that isn't yet available but which they are promising to have
available by the end of the year.) Most of this functionality can be defined
and configured to suit the user. So potentially the program could be used to
generate an extremely rich transcript.
What's missing from this is the ability to utilize the richness of XML
transcripts in QDA programs. The social sciences are really behind the curve
on integrating and utilizing XML in an initial input for QDA programs.
There are in fact lots of linguistic tools, included some very sophisticated
analysis tools, out there that can take Trancriber's XML transcripts and
import it (e.g. ONZE Miner, TASX-annotator, Nite-XML, Audiamus, etc). I
suspect if you start off with a Transcriber transcript you can get it into
lots of different linguistic tools, many of which are built round XML. So
it's a great tool for linguists but maybe only a great transcription tool
for social science research with lots of unexploited potential.
At the moment I don't think there are too many QDA programs built around
XML. Utilization of XML is pretty primitive compared to what's going on in
other types of software (e.g. general office profuctivity software). The
obvious exception is Atlas-ti which has some very sophisticated XML export
functions now. But there is no XML import although some import functionality
is now being promised in the next version. Not that the developer didn't
note the value of "value added input data" years ago. Imagine importing a
transcript and even before you start coding the QDA program is aware of all
the different speakers and their properties, and knows which speaker is
associated with which speech turn. And if you you had a multimedia engine
within the program you would also be able to utilize the time-aligned audio
data. Why leave the audio data behind or have to choose between audio or
transcript? (You can actually sort of keep the audio if you output the XML
into a document using HTML or SMIL and assuming your QDA program can
activate those tags but this is a kludge that's awkward and has lots
limitations.)
My 2 cents.
--
Alan Stockdale, Ph.D.
Center for Applied Ethics and Professional Practice
Education Development Center
55 Chapel Street, Newton, MA 02458-1060
|