Andreas, your query raises nice strategy issues about hierarchical
indexing. Worth answering on this list, though QSR Forum's the more
usual site for N4 queries, because hierarchical categorisation is often
raised in qual software debates as a "rigidity" issue. It's an option
in NUD*IST (and in recent versions of other qual software) used for many
different ways of sorting ideas, bringing concepts together,
representing the dimensions of an idea etc. Whilst some of these uses
(e.g. dimensionalising) work from the "top" down (when researchers e.g.
gather and recode material on a broader concept) it's more common for
them to happen "data up", the unconnected categories close to the data
being brought together as the researcher's understanding clarifies and
more abstract concepts develop. You wrote:
Our strategy, however, starts from a number of children (Free Nodes)
that are to become parents on a more and more abstract level. So, we are
looking for key categories on a rather theoretical level. Since we want
to achive this while working with the IndexSystem we cannot at first
create nodes close to the root. The top of the pyramid should finally
represent the result of our integration process but not the beginning.
Is there a feature in NUD*IST to create a Parent Node out of two (base)
nodes? And can this new Parent Node later have a Grand parent node
together with another node?
No problem, so long as you rethink the picture of the tree that you got
from NUD*IST's display. The "Root" is merely a virtual node,
representing the whole tree. So placing a node there is just saying it's
in, not that it is superior. All nodes other than the Root can be cut,
copied, attached, merged, and if they are at the "top level" they can be
moved to below any other node by any of those means. N4 doesn't of
course mind if you continue to work with Free Nodes, or if you have all
of your nodes strung out at the top level - but if you wish to group
them in hierarchies you'll need as in any hierarchical system to create
the "parent" first, then attach the "child". Just like placing a file on
the C drive, then moving it into a folder once you've made the folder.
Each time you want to bring a node or a subtree of nodes in below a more
abstract parent, create the parent, cut the subtree and attach.
Sometimes it takes a juggle to get the parent in the right place first
(as with the C drive's folders!), but it's swift.
Incidentally I've seen this process documented splendidly in audit
trails of NUD*IST projects - archive dated lists of nodes and they'll
show the development and usually simplification of the index system as
understanding develops.
hope this helps
Lyn
Prof. Lyn Richards,
Research Professor of Qualitative Methodology,
University of Western Sydney, Macarthur,
Adjunct Professor, International Institute for Qualitative Methodology.
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
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|