Yes, I agree with JSG. I've tried the solutions you've mentioned above Jennifer, and these seem to be set up to do analysis within a source or node, not between sources or nodes.
For example, in the data set I am working with, I have autocoded a series of journal articles and it would be helpful to be able to visualise the frequency and intensity of overlap between nodes. When any source is coded for "change" for example, it would be helpful to see how often sources are coded to other nodes. I've tried cluster analysis but haven't got it to quite do this yet.
I suppose what I am looking for is correlation or co-occurence of coding between nodes to understand not only the concepts/themes in the data but also the potential relationships between these and other project items...maybe there's a way to connect nodes with relationships and chart/vizualise these?
My first thought would be something like a Venn diagram or a variation of the tree-form cluster diagram where each of the branches of the tree is labelled (unless this exists and I just could not find it??), but maybe some sort of frequency or distribution curve of nodes in sources so I can see that most people talk about these issues together, but there are others that talk about different groups of issues. This makes me think a network diagram or explore diagram that combines with cluster analysis and queries any item in the project rather than just cases or nodes or attributes might be more likely to present what JSG and I have in mind (presuming that's similar). I've had a go at using the sociograms to do this briefly but haven't come up with anything workable yet.
JSG - what are your thoughts?
Hope this helps.