At 02/05/2003 10:19, you wrote:
>Hi all
>
>has anyone used qualitative software for thematic coding / conceptual
>mapping of a body of research evidence? I.E. I want to review the literature
>in a particular area but the research methods and theoretical persepctives
>used have been very disparate and I do not think standard systematic review
>techniques are best suited - I think a thematic analysis would be better and
>I wondered if anyone had found qualitative software useful in a similar
>situation? (and which software would be best suited?)
If you are looking for a way to uncover hidden patterns and relationship in
a large body of research, then it might be a good idea to look at content
analysis or text mining software, such as our own software WordStat. The
product has been used for research in library science (I may be able to dig
some references of published research in library science using
WordStat). We have also been working lately with people from the NIH for
the development of new visualization tools for what they call "literature
profiling". We will introduce in version 4 a new heatmap charting tool
which combines dual clustering and large frequency tables where frequency
are showed using color brightness. This tool, commonly used in genome
research has been found to be an excellent tool to uncover hidden patterns
in literature. If you are interested, I suggest you take a look at the
article below:
Chaussabel, D. & Sher, A. (2002) Mining microarray expression
data by literature profiling. Genome Biology 2002 3(10).
http://genomebiology.com/2002/3/10/research/0055/abstract
The correspondence plots and the multidimensional scaling maps already
available in WordStat 3.1 are also quite useful for finding such patterns.
The new heatmap feature will be introduced in the next version of WordStat
4.0 that should be release in a week or two (see
www.simstat.com/wordstat4.htm). This version will also offer other tools
that might be useful, such as case clustering and promixity charts (ability
to view in descending order references similar to a given one).
Normand Peladeau
Provalis Research
www.simstat.com
|