***** To join INSNA, visit http://www.insna.org *****
Ken,
I think it is fair to say that the question of how a sample graph relates to the population graph it is sampled from is a fairly controversial issue. Depending on what your position is, you might argue that the subgraph induced by the sample of ties and actors you have is a completely different animal from the graph for the entire population. This is also a question of what you mean by "major characteristics of network structure" and whether your analysis respects the fact that your sample is in fact a sample. These issues and illustrations of the information loss due to degeneration are treated in (and papers cited within)
Koskinen, J. H., Robins, G. L., and Pattison, P. E. (2010). Analysing Exponential Random Graph (p-star) Models with Missing Data Using Bayesian Data Augmentation. Statistical Methodology, Vol. 7(3), 366-384.
Also, check out the literature on sampling on graphs (that give you exact measures on how much information there is in the sample given some assumptions...) for example: (note also the first issue of Social Networks)
Frank, O (1979), Sampling and estimation in large social networks, Social Networks, 1 (1) 91-101.
Karlberg, M. (1998). Triad count estimation in digraphs. Journal of Mathematical Sociology,
23:99-126.
Cheers,
johan
________________________________________
From: Social Networks Discussion Forum [[log in to unmask]] On Behalf Of Ken Vance-Borland [[log in to unmask]]
Sent: 05 February 2011 00:18
To: [log in to unmask]
Subject: [SOCNET] Sample size for network structure
***** To join INSNA, visit http://www.insna.org *****
Dear SOCNET,
Are there published studies that investigate the portion of a study population that needs to be sampled in order to begin revealing the major characteristics of network structure?
We have in mind a population of several hundred nature conservation actors in a particular geographic region, but lack the funding to be able to survey all of them about their relationships.
Many thanks, -ken
Ken Vance-Borland, Executive Director
The Conservation Planning Institute
"To protect and restore biological diversity through innovative conservation planning
that is focused on effective implementation"
Tel:(541)231-7949 Email: [log in to unmask] Skype: ken.vance.borland
http://www.conservationplanninginstitute.org/
CPI is an IUCN member organization http://www.iucn.org/
_____________________________________________________________________ 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.
|