Print

Print


***** To join INSNA, visit http://www.insna.org *****
Dear Jordi,

You may want to review Steve Borgatti's paper published in Connections - "Structural holes: Unpacking Burt's redundancy measures." Here's a link:
http://www.analytictech.com/connections/v20%281%29/holes.htm

Best,

Corey


On Sun, Oct 27, 2013 at 5:21 PM, Jordi Comas <[log in to unmask]> wrote:
***** To join INSNA, visit http://www.insna.org *****
Hi-

Conceptually, a given node's amount of brokerage as measure by itsconstraint a la Burt's definitions seems the same to me as a clustering coefficient.

In other words, for a given node, the constraint may be .5.  For the whole network, those constraint measures could be averaged.

A weighted clustering coefficient would measure for each node, how much triadic closure is observed in its one-degree network neighborhood.

Thoughts?  If there are differences in measures, is it due to networks perhaps having several components? Or isolates?

A quick check using UCINET on a network I study shows that constraint is .89 and clustering is .27; the "desnity" measure the structural holes routine generates is the same as the clustering coefficient.

But my basic question is if one wants to analyze the amount of local closure or clustering, is looking at average constraint and cluster coefficient a bit redundant?

Thanks for any insight!

Jordi

--
Jordi Comas

"There is nothing so practical as a good theory."  Kurt Lewin

Assistant Professor
School of Management
Bucknell University
Taylor 112
570 577 3161

Research and Writing Blog: Nets We Weave


_____________________________________________________________________ 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.



--
Corey Phelps, Ph.D.
Associate Professor
Department of Strategy and Business Policy
HEC Paris
1, rue de la Libération
78351 Jouy-en-Josas
France
Phone: (33) 1 39 67 74 15
Fax: (33) 1 39 67 70 84
http://www.hec.edu/Faculty-Research/Faculty-Directory/PHELPS-Corey-C
http://ssrn.com/author=341753
http://fr.linkedin.com/pub/corey-phelps/5/53b/782/
_____________________________________________________________________ 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.