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


For what it's worth, I think this paper is awful and should never be
cited...   :-)

There is no one right way to normalize. Depends on what you are trying
to measure.  The logic of normalizing by ego is that you want to remove
individual differences in the average tie strength and just focusing on
structure.  If you normalize based on the entire network, part of what
drives density is variation in average tie strength.  This strategy for
normalization is implemented in STRUCTURE; definitely not original to
us.  And, BTW, our results don't depend on the normalization.


Ezra Zuckerman

-----Original Message-----
From: Social Networks Discussion Forum [mailto:[log in to unmask]] On
Behalf Of Doug Bryan
Sent: Friday, January 17, 2003 1:56 PM
To: [log in to unmask]
Subject: Reagans' & Zuckerman's team density metric


I have 2 questions regarding Reagans and Zuckerman's work on R&D teams
(Reagans R and Zuckerman E (2001) "Networks, diversity and productivity:
social capital of corporate R&D teams," Organization Science,

1. Is the paper highly regarded?  i.e., is it often cited?

2. In their metric for "team density" they normalize the strength of
from i to j using the strength of the strongest tie from i.  (See
figure.)  Why?  Can people comment on the pros/cons of normalizing
to each individual (max(tie(i,*))) versus normalizing relative to the
team (max(tie(*,*))).

Doug Bryan
[log in to unmask]

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.