Print

Print


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

You mean by weighting/coloring the alters according to their different attributes?

If so, then might not I be assuming into my "IV" what should naturally be the "DV"?
What kind of attributes are you suggesting? 

Of if that is not what you had in mind, then won't there be a serious problem of missing data for all those relations that might have been reported by people who were not in the sample? 



Btw, my answers to John's questions are that we should assume that the relations are asymmetric or "indeterminate", and that there is very little overlap.

Blyden
 


-----Original Message-----
From: [log in to unmask] [mailto:[log in to unmask]] 
Sent: Friday, June 05, 2009 5:26 PM
To: [log in to unmask]
Cc: [log in to unmask]; [log in to unmask]
Subject: Re: equivalence from egonets?

Blyden, I missed your answers to these questions.  Nonetheless, one approach is to sum the sets of directed links in each binary relation and compute structural equivalence on the result, if the individuals have attributes.  Paul Broome

From: "John P. Boyd" 
Date: 30/05/09 22:41
Subject: Re: equivalence from egonets?


Blyden, how much overlap in the ego networks is there? If there is none, there is not much you can say. Is the relation symmetric, like "Sibling", antisymmetric, like "Father", or indeterminate, like "Friend"?yours, John  PS A note on terminology: one says "equivalence relation" for the relation and "equivalence classes" for the induced sets. I've also seen "blocks" as in "blockmodeling" for the equivalence classes. But "equivalence set" is very jarring. 

----- Original Message ----- From:Blyden PottsTo:[log in to unmask]:Saturday, May 30, 2009 4:14 AM
Subject:equivalence from egonets?
I?d like to get other network analysts? suggestions on how they would handle this problem: 
 
How would you best assign individuals from a sample to partitions (i.e. equivalence sets) in a network when the only 
info you have about the structure of the overall network is the ego networks of the sample persons?  How can you 
best estimate the equivalence sets of the overall network from this limited data?
 
Blyden Potts
 _____________________________________________________________________ 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.