Mitchell Centre seminar series
| Dates: | 2 March 2022 |
|---|
| Times: | 16:00 - 17:00 |
|---|
| What is it: | Seminar |
|---|
| Organiser: | School of Social Sciences |
|---|
|
Michal Bojanowski
Kozminski University
Exponential-family Random Graph Models for egocentrically-sampled data: examples and new developments with illustrations using data on core discussion networks
Egocentric sampling of networks selects a subset of nodes ("egos") and collects information from them on themselves and their immediate network neighbours ("alters"), leaving the rest of the nodes in the network unobserved. This design is popular because it is relatively inexpensive to implement and can be integrated into standard sample surveys. Recent work has shown that data collected through an egocentric design can be used to estimate certain specifications of Exponential-family Random Graph Models (ERGM). In particular, data about the egos’ immediate connections can be used to estimate models with nodal, degree, and mixing effects. This talk will first review the proposed modeling approach focusing on (1) survey sampling design and measurement considerations, (2) types of model terms that are identifiable, and (3) properties of the model in its ability to recover whole-network features. It will then proceed to present results of a new re-analysis of GSS 2004 data with a question about the effect of marriage ties on the structure of core discussion networks - an analysis showcasing how to estimate models with effects of dyadic covariates which are *not* functions of observed node attributes. The talk will conclude with a discussion of advantages and limitations of the presented approach.
Travel and Contact Information
Find event
G6
Humanities Bridgeford Street
Manchester
