Explorations in Structural Analysis (RLE Social Theory) by Ronald L. Breiger

Author:Ronald L. Breiger [Breiger, Ronald L.]
Language: eng
Format: epub
Tags: Social Science, Sociology, General
ISBN: 9781317657583
Google: 5AhUBAAAQBAJ
Publisher: Routledge
Published: 2014-08-21T05:00:38+00:00

+ Entries in Table 7 were computed from six tables each of the form

Table 8. Intersection of CONCOR partition and attribute partition, for pairs of blocks: Eight-block level.+

+See notes to Table 7.

expect to locate where the mismatches of network and attribute structures are concentrated. The proportion of agreement between the CONCOR partition and the attribute partition is reported above the diagonal of Tables 7 and 8, and the probability of at least this much agreement is reported below the diagonal. At the four-block level the mismatch is concentrated in block (II) with respect to block (IV). At the eight-block level, agreement is better than 90% for 15 of the 28 comparisons; correspondingly, the mismatch is severe in 5 of the 28. The discrimination among a trio of blocks (C, D, and H) is poor (less than 50% in each case). The main conclusion is that an attribute analysis would differ from the network analysis in grouping the “outsiders” (H) with the other two blocks, which are themselves grouped together in the four-block and two-block CONCOR splits. At the same time, the block memberships, which cohere in a relational structure, are also seen to cohere by the similarity of their attributes: 20 of the 28 comparisons are significant at the .001 level (Table 8), indicating that block members are usually more similar internally (on the two metric attributes) than they are to members of a given other block.

CONCLUSION

Blockmodel analysis provides a general approach for the study of social structures through a consideration of multiple networks. This paper has stressed the identification of highly idealized patterns of interaction from the complex interweaving of thousands of pair relationships.

Within the sociology of science, blockmodel analysis may be viewed as a means for extending and giving an operational meaning to the qualitative insight of Menzel (1962) that interactions which often seem to occur “by chance” on the micro level cumulate into aggregate regularities. The scientists studied here were seen to maintain a stable structure of contact and unawareness choices. By examining a block’s “status” as a position in a pattern among the positions, rather than as a rank attribute,21 the identification of a stratification system, as well as the identification of the boundaries of and discontinuities within that system, could be unified in a single analysis. Blockmodel analysis thus seems particularly useful in applications to populations whose boundaries are “open” and whose membership bridges conventional categorical positions (e.g., scientific disciplines, academic ranks).

Interpretation of the structure of social relationships was found to be enriched, not confused or diluted, when the career attributes of block members were brought into the picture. However, substantial analytical leverage was gained by taking the relational structure itself as the primary object of study, leaving the articulation of this structure with individual attributes as a problematic issue requiring elucidation. One example of this leverage was the linking of stratification to structural awareness, a linkage which proved elusive in previous studies (Cole and Cole, 1968; 1973). Another example of this increased leverage was

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.