Indicators of collaboration intensity?

[Steven A. Morris]

Hi folks,

Does anyone out there know of an indicator of the "intensity" of
collaboration that characterizes a research specialty?  For example,
mathematicians tend to be loners that don't often work with others,
therefore mathematical specialties have "low" collaboration intensity.
Biomedical researchers tend to work in large teams, so biomedical
specialties have "high" collaboration intensity.

One possible indicator would be the mean authors per paper in a
specialty's literature. This is actually not a very sensitive indicator I
think.  Another indicator could be the percentage of single authored
papers, which I think is more sensitive, e.g.,

Specialty   percent 1     collaboration
            auth papers   intensity
======================================
Distance     47%           low
eductation

Complex      23%           medium
networks

angio-       7%            high
 genesis

Does anyone have any ideas or references on
collaboration intensity indicators?

Thanks,

Steven Morris
Oklahoma State University