TY - JOUR
T1 - Macrostructure from Microstructure
T2 - Generating Whole Systems from Ego Networks
AU - Smith, Jeffrey A.
N1 - Funding Information:
The author disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was partially supported by NSF:HSD 0624158 and NIH: 1R21HD068317-01.
Funding Information:
The key questions in this paper emerged out of discussion with Miller McPherson. James Moody, Lynn Smith-Lovin, Robin Gauthier, and the network working group at Duke University also provided helpful feedback. The author would also like to thank James Moody for providing the code and assistance to create the Sociology Coauthorship network used in this paper. Parts of this paper were presented at the 2011 Sunbelt Social Networks Conference. This research uses data from Add Health, a program project designed by J. Richard Udry, Peter S. Bearman, and Kathleen Mullan Harris, and funded by grant P01-HD31921 from the Eunice Kennedy Shriver National Institute of Child Health and Human Development, with cooperative funding from 17 other agencies. Special acknowledgment is due Ronald R. Rindfuss and Barbara Entwisle for assistance in the original design. Persons interested in obtaining data files from Add Health should contact Add Health, Carolina Population Center, 123 W. Franklin Street, Chapel Hill NC 27516-2524 ( [email protected] ). No direct support was received from grant P01-HD31921 for this analysis. Direct correspondence to Jeffrey A. Smith at [email protected] .
PY - 2012/8
Y1 - 2012/8
N2 - This article presents a new simulation method to make global network inference from sampled data. The proposed method takes sampled ego network data and uses exponential random graph models (ERGM) to reconstruct the features of the true, unknown network. After describing the method, the author presents two validity checks of the approach: the first uses the 20 largest Add Health networks while the second uses the Sociology Coauthorship network in the 1990s. For each test, I take random ego network samples from the known networks and use my method to make global network inference. The method successfully reproduces the properties of the networks, such as distance and main component size. The results also suggest that simpler, baseline models provide considerably worse estimates for most network properties. The paper concludes with a discussion of the bounds/limitations of ego network sampling as well as possible extensions to the proposed approach.
AB - This article presents a new simulation method to make global network inference from sampled data. The proposed method takes sampled ego network data and uses exponential random graph models (ERGM) to reconstruct the features of the true, unknown network. After describing the method, the author presents two validity checks of the approach: the first uses the 20 largest Add Health networks while the second uses the Sociology Coauthorship network in the 1990s. For each test, I take random ego network samples from the known networks and use my method to make global network inference. The method successfully reproduces the properties of the networks, such as distance and main component size. The results also suggest that simpler, baseline models provide considerably worse estimates for most network properties. The paper concludes with a discussion of the bounds/limitations of ego network sampling as well as possible extensions to the proposed approach.
KW - ERGM
KW - Networks
KW - sampling
KW - simulation
UR - http://www.scopus.com/inward/record.url?scp=84993768037&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84993768037&partnerID=8YFLogxK
U2 - 10.1177/0081175012455628
DO - 10.1177/0081175012455628
M3 - Article
C2 - 25339783
AN - SCOPUS:84993768037
SN - 0081-1750
VL - 42
SP - 155
EP - 205
JO - Sociological Methodology
JF - Sociological Methodology
IS - 1
ER -