Using log linear analysis with network data: another look at sampson's monastery

An explanation, in sociologically significant terms, of Stephen E. Fienberg's & Stanley S. Wasserman's method (see SA 31:4/83N2144) for applying loglinear analysis to social network data. The method is applied to a well-analyzed data set reported by S. F. Sampson (Crises in a Cloister,...

Full description

Saved in:
Bibliographic Details
Published in:Social networks 1982-09, Vol.4 (3), p.243-256
Main Author: Reitz, Karl P.
Format: Article
Language:English
Subjects:
Social network/Social networks/Social networking
Statistics/Statistical/ Statistically
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:An explanation, in sociologically significant terms, of Stephen E. Fienberg's & Stanley S. Wasserman's method (see SA 31:4/83N2144) for applying loglinear analysis to social network data. The method is applied to a well-analyzed data set reported by S. F. Sampson (Crises in a Cloister, PhD dissertation, Cornell U, 1969) consisting of the network of 8 relations on a group of 18 novices in a monastery. Here, the union of the 4 positive relations is formed & used as the basis for analysis. The model that best fits the data is shown to be the one that hypothesizes individual parameters of attractiveness, expansiveness, & the tendency to reciprocate. These parameters are all network derived individual attributes. No second order terms are necessary to explain the data. These parameters are interpreted & used to explain the events as reported by Sampson. 4 Tables, 8 References. AA.
ISSN:0378-8733
1879-2111
DOI:10.1016/0378-8733(82)90025-9