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

Current versions of UCINET offer the option to "compute informant  
accuracy" when examining cognitive social structure (CSS) data.  As  
the online documentation  
(https://sites.google.com/site/ucinetsoftware/document/faq/accuracymeasuresincssroutine) explains: "These measure the agreement between an individual's perception of the network (their "slice") with the "true"  
network."

However, the documentation does not say what is used as the "true"  
network against which informants' reports are compared for accuracy.   
I suspect UCINET uses some version of a locally aggregated structure  
as the ground truth, but that's just a guess.  Any thoughts on what  
UCINET is doing here would be appreciated.

Along the same lines - if you have any suggestions for past work on  
measuring informant accuracy within CSS-type data, they would be  
appreciated as well.

Thanks,
Zachary

-- 
Zachary Neal, Assistant Professor
Michigan State University
Department of Psychology & Global Urban Studies Program
316 Physics Road, Rm 262
East Lansing, MI 48824
http://www.msu.edu/~zpneal
Twitter: @zpneal

http://www.routledge.com/books/details/9780415881425/
http://www.routledge.com/books/details/9780415997270/
http://www.youtube.com/watch?v=Az6Uy7p3kas

_____________________________________________________________________
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.