An interesting question Louise. I don’t’ think I’ve seen more than three in any project, and certainly I have not used more than three myself.
I wonder, however, if the over all number of nodes and the structure of the project is of more concern than the number of levels?
I use NVivo, and if I manage to get a treed structure with material that is coded for analytic purposes rather than simply indexed by topic I feel very pleased. Categories, subcategories and perhaps sub sub categories create a structure in my head and on the screen. If I’ve got clear trees, and not too many of them, I can display the structure quite well in the list view of NVivo7. So if I’m coding I have my data open on the detail view and at the side I have my trees, with all the sub categories and sub sub categories visible. I can drag and drop for coding. It feels rather like sorting the laundry into a well organised wardrobe and the complexity doesn't matter. If I’m thinking about the project, I can look at the nodes and work out what I might want to ask next. A node structure fits well with drawing models.
But if I have nothing but lots of free nodes, I find myself looking at a bit of data, creating a new node, then deciding that an existing node will do, checking the node properties for the definition, and then fretfully re-creating the new node* I don’t have a vision of where I’m going. It’s as if I didn’t have a wardrobe, , so I can’t tell where to put the laundry, and I know that tomorrow morning I won’t be able to find any clean socks. I look at the free nodes and can’t think what I can ask about them, or how they might hang together in a model*
Of course I often need that first phase of free nodes and messy complexity when coding up from data, but because it’s daunting, I try to limit the number of free nodes, and where possible to move my free nodes into orderly trees (building wardrobes I guess).
The laundry metaphor was coined by Leonie Dawes and Pat Bazeley uses it in her recent book 'Qualitative Data analysis with NVivo' (Sage 2007) She also discusses lumping and splitting very helpfully, pointing out they are very much a matter of personal preference. I’d love to know what other people think about the issue of complexity.
Hope this helps
Helen Marshall
Dr Helen Marshall
Senior Associate, Centre for Applied Social Research
Royal Melbourne Institute of Technology
GPO Box 2476V Melbourne 3001
Victoria Australia
ph 61 3 99259073
fax 61 3 9925 3088
email [log in to unmask]
"The best cure for being sad...is to learn something" (T.H White 'The Once and Future King')
>>> "Corti, Louise" <[log in to unmask]> 22/08/2007 23:06 >>>
Hi all,
I'm looking for an indication of the extent to which the average
researcher creates nested codes. How complex do coding trees usually
get? How many levels deep do codes typically be nested? Personally, Ive
never gone more than about 3 levels deep as I find it becomes too
complex.
Thanks!
Louise
Louise Corti
Associate Director & Head ESDS Qualidata, Outreach & Training
UK Data Archive - a service provider of the Economic and Social Data
Service (ESDS)
University of Essex
Colchester CO4 3SQ UK
Tel: +44 (0) 1206 872145
Email: [log in to unmask] <mailto:[log in to unmask]>
Web: http://www.data-archive.ac.uk <http://www.data-archive.ac.uk/> and
www.esds.ac.uk <http://www.esds.ac.uk/>
**********************************************************************
Legal Disclaimer: Any views expressed by the sender of this message are
not necessarily those of the UK Data Archive or the ESDS. This Email and
any files transmitted with it are confidential and intended solely for
the use of the individual(s) or entity to whom they are addressed.
**********************************************************************
|