Print

Print


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

Qian

Yes we reported the apparent robustness of certain centrality measures under
conditions of missing data in:

Costenbader, E., & Valente T.W. (2003). The stability of centrality measures
when networks are sampled.  Social Networks, 25, 283-307.

And in-degree centrality is quite robust to missing data.

-Tom

 

From: Social Networks Discussion Forum [mailto:[log in to unmask]] On
Behalf Of Qian Hu
Sent: Monday, February 11, 2013 7:56 PM
To: [log in to unmask]
Subject: Missing network data

 

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

Dear SOCNET members,

I have two questions about the missing network data. 

what can be done if we cannot get an ideal response rate (say less than 70
percent, or even lower)?

1.	I remember somewhere in the literature that says some centrality
measures such as the indegree centrality, are relatively robust for network
data that captures over 50% of the data. Am I correct?
2.	The existing research seems to suggest two methods to deal with
missing data: exponential random graph models, and imputation methods. If we
know the key actors are captured in the imperfect data set. Can we leave
them as they are without using any additional modeling process? Any
literature to support this argument?

 

 

 

Thank you very much. 

Qian Hu

University of Central Florida 

_____________________________________________________________________ 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. 


_____________________________________________________________________
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.