*****  To join INSNA, visit  *****

Hi list,

So I received 'Scott. "Social Network Analysis: A Handbook"'  and
Wasserman and Faust and have been reading through them (and finding it
weird to see people's names who I know are on this list written in
print :)). As I do I find questions to ask. The first one is about how
I notice that a lot of the measures Scott describes require an
arbitrary pick for the cut off threshold between what to count and
what not to count (e.g. n-cliques, k-plexes, m-clans). At first
glance, this seems prone to error and to potentially resulting in
greatly different analyses merely by tweaking a parameter. My question
is has there been any work toward making sort of gradient measures,
where all possible values of the parameter are averaged over? Then in
one measure there is one single way to determine cliques or whatever
you're trying to measures; or perhaps there is a fundamental reason
why this is impossible?

Thanks in advance,
-Nick Guenther

SOCNET is a service of INSNA, the professional association for social
network researchers ( To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.