I was looking for a very long time to find a good workflow with regards to processing transcribed interviews with codes and keywords. I checked out all kinds of software, but I prefer to keep all my research and writing in one place: Scrivener. I am writing a PhD and using quite a few long interviews, which I will need to code to make them more easy to work with.
Here is the workflow I devised:
-
Transcribe with Transcriva http://www.bartastechnologies.com/products/transcriva
A tip is to break the interview down into smaller chunks as you transcribe. So if the interviewee talks for a longer time, hit return to start a new timed section. This makes it easier for you to jump to that exact place in the audio recording, and will also make the coding in Scrivener easier later.
2. Export the interview as RTF. Include the time codes. Open in your text editor. I use Nisus Writer Pro, but anything will do.
Before and after each interviewer/interviewee name place a ‘#’. You can do this by using ‘Replace and Find’. So if your interviewee is called ‘John Do’, you search for John Do and replace with # John Do #. Make sure there is a double return after each name (this should happen automatically when you export from Transcriva anyway). This is also a good time to change your interviewee names if you want them to be anonymous. So for example change to # Person 1 # instead of # John Do #.
After doing this save as an RTF
-
Import file as Multimarkdown in Scrivener (see elsewhere on this forum how to install the Multimarkdown plugin http://www.literatureandlatte.com/forum/viewforum.php?f=21). This will import your interview into a folder with a separate file for each question and answer.
-
Code interview. This now gives you a number of different coding options. For example if you look at the folder in ‘Outliner’ mode you can assign each snippet with a different theme by changing ‘Label’ into ‘Theme’, and adding different themes to suit you. The same can be done with ‘Status’, for example into ‘Context’ and add different contexts to suit the interview. You can then order the snippets according to these themes and contexts.
By clicking on each file, you can assign keywords to each question and answer. You can highlight parts of the text with different colours, annotate within the text, or add notes. All of which are searchable.
You can also add links or images relating to what your interviewee is talking about.
One more handy thing is to click on the folder containing your interview, under the Documents menu go to convert/convert to file. Now you can press Cmd-Opt-1 or go to Documents menu Edit Scrivenings/All Content to view your whole interview.
I have yet to try this out extensively as I just came up with it last night, but I think it could work as a way of using Scrivener for qualitative data. Would be interested to hear whether other people have found other methods, or if they have any luck with this one.
Thanks! Lou