Social Network Analysis: State of the art and new developments
Tuesday 15/3/11, 2-6.30pm at the Royal Statistical Society, 12 Errol Street, London, EC1Y 8LX.
Social network analysis has grown from being a small niche curiosity to a well respected methodology in a relatively short time. This half-day meeting provides an overview of the field and its applications, and a discussion of recent developments. Topics covered include ideas, concepts and the history of the network perspective; multilevel models to analyse network and group dependencies, for example in educational data; models for dynamics in longitudinally observed networks and their use to study peer influence; using Exponential Random Graph models to explore network structures; and specialist software available for analysing network data.
Martin G. Everett (University of Manchester)
Social Network Analysis: History and Overview
- Ideas, concepts and history of the network perspective.
- Techniques for visualization and descriptive measures.
- Specialist software available for analyzing network data.
Mark Tranmer (University of Manchester)
Random effects models for social network and group dependencies
- The "network effects model" and the "network disturbances model".
- Multilevel random effects models for modeling group dependencies.
- Assessing variation in educational outcomes between pupils and schools, taking into account friendship networks of the pupils in the school.
Tom A.B. Snijders (Nuffield College, University of Oxford)
Modelling dynamics of social networks
- Stochastic actor-oriented models: a class of continuous-time Markov chain models for representing network dynamics.
- Models of network dynamics for studying peer influence.
- How individual behaviours (adolescent smoking and collaboration between firms) are influenced by their network partners (“peers”).
Johan Koskinen (University of Manchester)
Recent advances in Exponential Random Graph Modelling (ERGM) and inference
- Using ERGM to explore network structures
- Estimating network size and handling heterogeneity in ERGM
- Different approaches to Bayesian estimation for ERGM
Registration with payment is required. Booking forms can be downloaded from http://membership.rss.org.uk/main.asp?group=&page=1321&event=1247&month=3&year=2011&date=
or requested by email from [log in to unmask]
Meeting contacts: Tarani Chandola ([log in to unmask]) and Jouni Kuha ([log in to unmask])
Registration charges: RSS Student/Retired Fellows - £20;
CStats/GradStats - £22;
RSS Fellows - £25; RSS members - £30; none of the above - £35