SOCn5010 Analýza sociálních sítí
Přednáška 12: Statistické nástroje


Type of hypotheses
•Node-level (monadic) – cases are nodes – higher centrality in professional network – higher wage
•Dyadic level – cases are pairs - the stronger the tie of professional cooperation, the stronger
the tie of mutual trust
•Group/network level – the higher the density of the network, the faster the spread of innovation
•
•

Statistics
•Description of a network
•Hypotheses about theoretical parameter
•Hypotheses about two paired means/densities (test for differences in the probability of a tie of
one type and the probability of a tie of another type)
•Correlation between two networks with the same actors (if there is a tie of one type among two
actors, is there a likelihood of a tie of another type)

Node-level regression
•Symmetric associations (correlation) Vs. Assymetric relations (regression)
•Regressing position on attributes
•Attributes explaining the position of node
•Attributes measured at interval level
•E.g. Predicting centrality in a friendship network using age and income variables

Dyadic regression
•Predicting a relation from another one
•Dependent network, independent network + node attributes, regressing each element in dependent
network on its corresponding elements in the independent network + attribute-similarity network
•E.g. Predicting friedship relation by co-occurence network (attendance of same uni courses) +
gender

References
•HANNEMAN, Robert A. and Mark RIDDLE.  2005.  Introduction to social network methods.  Riverside,
CA:  University of California.
•BORGATTI, Stephen P., Martin G. EVERETT and Jeffrey C. JOHNSON. 2013. Analyzing Social Networks.
Thousand Oaks, CA: Sage
•CROSSLEY, Nick. 2010. Towards Relational Sociology. Abingdon: Routledge.
•PRELL, Christine. 2012. Social Network Analysis: History, Theory & Methodology. Los Angeles: Sage.
•KNOKE, David, and Song YANG. 2008. Social network analysis. Thousand Oaks: Sage.
•