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

Hi. For those of you working on personal networks, using EgoNet or 
thinking of using it, we worked with software author MdLogix recently to 
update the software and add a randomization feature that will allow the 
program to randomly sample the list of alters originally solicited. 

Research shows that personal network structural measures remain similar 
for randomly generated subsamples (with a minimum n, depending on which 
measure, as few as 12 for some, as many as 25 for others) of the orginal 
personal network (of approximately 45 individuals). 

This feature will decrease informant burden by more than half when using 
the program to elicit networks, especially if you're interested in 
personal network structure and not needing all of the ties. Unfortunately, 
I don't think it will make it any easier to conduct methodological 
research on the effect of sampling on personal networks, because once 
you've collected reduced amounts of data, you don't have the data for each 
individual in each personal network from which to sample--you've only got 
the already sampled (and smaller) network.


Eric C. Jones, Ph.D.
Department of Anthropology
University of North Carolina at Greensboro
426 Graham Building
Greensboro NC 27412-5001
office: 336-334-5132
[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.