1 / 33

Finding Social Groups: A Meta-Analysis of the Southern Women Data Linton C. Freeman

Photograph by Ben Shahn, Natchez, MS, October, 1935. Finding Social Groups: A Meta-Analysis of the Southern Women Data Linton C. Freeman.

colum
Download Presentation

Finding Social Groups: A Meta-Analysis of the Southern Women Data Linton C. Freeman

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Photograph by Ben Shahn, Natchez, MS, October, 1935 Finding Social Groups: A Meta-Analysis of the Southern Women Data Linton C. Freeman

  2. In 1933 W. Lloyd Warner was teaching at Harvard. He decided to send four graduate students, Allison and Elizabeth Davis and Burleigh and Mary Gardner to study race and social class in Natchez, Mississippi.They collected systematic two mode data on the participation of 18 women in 14 small informal social events.

  3. p.148

  4. Davis, Gardner and Gardner sought: 1. To specify tightly knit groups 2. To assign women to core and peripheral positions in their assigned groups They said:

  5. Where it is evident that a group of people participate together in these informal activities consistently, it is obvious that a clique had been isolated. Interviewing can then be used to clarify the relationship. Those individuals who participate together most often and at the most intimate affairs are called core members; those who participate with core members upon some occasions but never as a group by themselves alone are called primary members; while individuals on the fringes, who participate only infrequently, constitute the secondary members of a clique. p. 150

  6. DGG described the groups they came up with: Women 1-9 in one group 9-18 in the other group Woman 9 in both groupsAnd they specified positions in each:1-4 & 13-15 Core 5-7 & 11-12 Primary 8-9 & 9,10, 16, 17, 18 Secondary

  7. DGG described the groups they saw: Women 1-9 in one group 9-18 in the other group Woman 9 in both groupsAnd they specified positions in each:1-4 & 13-15 Core 5-7 & 11-12 Primary 8-9 & 9,10, 16, 17, 18 Secondary

  8. Since then: 21 procedures have been used to assign women to groups, and 11 to assign positions in the groups They are:

  9. DGG 1941 Intuition Homans 1951 Intuition Phillips and Conviser 1972 Information Theory Breiger 1974 Matrix Algebra Breiger, Boorman & Arabie 1975 Computational Bonacich 1978 Boolean Algebra Doreian 1979 Algebraic Topology Bonacich 1991 Correspondence Analysis Freeman 1992 G-Transitivity Everett & Borgatti 1993 Regular Coloring Freeman 1993 Genetic Algorithm I & II Freeman & White 1993 Galois Lattices I & II Borgatti & Everett 1997 Bipartite Analyses I, II & III Skvoretz & Faust 1999 p* Model Roberts 2000 Normalized SVD Osbourn 2000 VERI Procedure Newman 2001 Weighted Proximities

  10. And they assigned women to groups:

  11. Group Assignments

  12. And they assigned positions:

  13. Core/Periphery Assignments

  14. Here I will do a kind of meta-analysis: one data set several analytic procedures. Schmid, Koch, and LaVange (1991): Meta-analysis is “. . . a statistical analysis of the data from some collection of studies in order to synthesize the results.”

  15. The Question of Group Membership

  16. The Question of Group Membership Batchelder, Romney and Weller—Consensus Analysis Gets: “true” answers (consensus) “competence of judges” (approach to consensus)

  17. (Based on iteration to maximum likelihood)

  18. Then calculate matches and covariance. If they agree, factor analyze and the first factor estimates “competence.”

  19. Then calculate matches and covariance. If they agree, factor analyze and the first factor estimates “competence.” Here the correlation between matches and covariance is .967

  20. The Question of Core and Periphery

  21. Two Methods for Ordering: Gower’s (1977) canonical analysis of asymmetry (algebraic-deterministic) Batchelder and Bershad’s (1979) dynamic paired-comparison scaling (probabilistic)

  22. Here, I’ll try to interpret dimensions 2 and 3 of the principal components analysis. Here they are:

  23. They show a consistent pattern in terms of the way they depart from the consensual pattern:

  24. Through time there has been a very slow, but steady movement toward the consensual pattern

  25. And, we can evaluate the several families of approaches to uncovering groups:

  26. Procedure N Average Score Statistical model 1 .957 Eigen structure 3 .954 Optimal partition 5 .941 Transitivity 1 .926 Cliques 1 .916 Algebraic duality 6 .914 Intuition 2 .887

More Related