***** To join INSNA, visit http://www.insna.org *****
Not systematic, but I¹ve found that beyond 7 pages it gets dodgy‹I think
it depends more on how motivated the respondents are through an inherent
interest in the list that you give them.
Here¹s a version of our rosters used in Kenya, Ghana, and India for a
couple of decades.
http://worldsci.net/ee/index.php/pages/main/6
(look for the instruments marked ³network²)
Wes
Wesley Shrum
Professor of Sociology, Louisiana State University
Director, Ethnografilm <http://ethografilm.com>, 8-12 April 2015;
Director, Fringe Performance Archive <http:fringearchive.org>, National
Library of Scotland
Program Officer, Society for Social Studies of Science
<http://4sonline.org>
------------------------------
Date: Tue, 4 Nov 2014 13:31:25 -0500
From: "< Janet Okamoto >" <[log in to unmask]>
Subject: Network roster size and network data quality
***** To join INSNA, visit http://www.insna.org <http://www.insna.org/>
*****
Hi all,
I'm currently working on a project collecting network information from a
community
health coalition and am looking for any scholarly references to work that
has examined
optimal size or thresholds of network rosters given to survey respondents.
Is there any
work that has looked at what size or at which point the quality of your
data starts to
decrease? How many names and relationships can a person cognitively
process at one
time before fatigue sets in and you have to start worrying about data
quality? Measurement, meaning, and psychometric properties of network
survey questions has
been an interest of mine for years, but I haven't seen many publications
about the topic,
so am hoping some on this list can help.
Thanks,
Janet
Janet Okamoto, Ph.D
Research Associate
Mayo Clinic
[log in to unmask]
_____________________________________________________________________
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.
|