***** To join INSNA, visit http://www.insna.org *****
Hi all -
I've got a question for the group on two means of identifying subgroups in
I'm trying to come up with a variable I can use to identify which
subgroup(s) my networks' nodes belong to. I've got 23 2-mode networks,
the smallest of which is 269 x 437, the largest of which is about 4500 x
20000 (twenty thousand).
After I transformed my two-mode matrices into 1-mode matrices, I tried
using UCINET's "Lambda sets" option, as the description in Wasserman &
Faust's book (Social Network Analysis, 1994) made the most intuitive sense
to me (a subgroup is composed of nodes whose ties to each other are more
frequent that their ties to "outsiders"). I also tried Pajek's
"k-components" option, based on what I read in Moody & White's 2003 ASR
article and the description in the Pajek manual (Exploratory Social
Network Analysis with Pajek, 2005).
My understanding is that the Lambda set measure is based on edge
connectivity while the k-component measure is based on node connectivity.
However, I don't claim to know the underlying equations to calculate each.
The issue I have is that, even though these don't seem like the same
measure, I get essentially the same results when graphing them (I've only
playing around with the smallest of my networks; I figured I'd better check
the experts before proceeding with the rest!). So my question is, does it
make sense that the results (the assignment of nodes into subgroups) would
be the same for both measures, and if not, what might the similarity of
results be due to?
I'm grateful for any advice from the group, and thank you in advance for
Jennifer van Stelle
Department of Sociology
Stanford, California 94305-2047
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.insna.org). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.