3K Advanced Social Network Analysis II – Influence Mechanisms & Longitudinal Network Analysis
Filip Agneessens, University of Groningen
6 - 17 August (two week course / 35 hrs)
Detailed Course Outline [PDF]
Course Content
This course covers advanced statistical methods for analyzing social network data, focusing on testing hypotheses about network effects on individual attributes (social influence or contagion models), network structure of groups on group outcomes, and models to disentangle selection from influence. We begin with statistical models for the effect of structural position of actors in a network (centrality, structural holes, closure, diversity, range, Simmelian ties, …), and then move onto statistical models where actors adjust their own attitudes and behaviour to that of others they are connected with (attitudinal and behavioural contagion). In week 2 we then move to longitudinal models for influence, as well as methods to simultaneously model selection and influence processes. Topics include QAP regression, (longitudinal) autocorrelation models and stochastic models for dynamic network analysis (SIENA). We also examine statistical work that incorporates both individual and group level effects using multilevel models. Software used includes: the “spdep” package in R, and SIENA in Stocnet. Additional programs for R will be made available throughout the course.
Course Objectives
The course aims to familiarize participants with the formal statistical analysis of network data for influence mechanisms. Participants will become familiar with specific programmes designed for these analyses and with the mathematical basis for the modelling approaches, and they will learn how to conduct statistical analyses of their own network data. Participants are encouraged to bring with them their own network data to be analyzed using the techniques covered.
In general the course focuses on 2 aspects:
1) how does the position or structure of the network influence individual actor (or organisational) characteristics, such as attitudes and behaviour, as well as how the structure as a whole will have an impact on individual or group outcomes?
Participants should be able to answer questions, such as:
- Are actors who are more central more likely to have a high performance?
- Are brokers in a network more likely to move up in an organisation?
- Are pupils influenced in their smoking behaviour by the smoking behaviour of their friends?
2) Whether the above social influence mechanisms can be detected over time or whether selection mechanisms, based on for example homophily or transitivity are the major sources of changes in network configuration. More specifically, this will enable participants to answer questions, such as:
- Do employees trust others if they have a similar attitude towards work, or do their attitudes change as a result of their position in the trust network?
- Does having criminal friends make students more likely to become criminal as well, or do students have a similar tendency to criminal behaviour tend to develop friendships?
- Does transitivity in an advice network occur because employees start going for advice to those other employees, who go to the same third parties for advice? Or does it occur because they start going for advice to those that their advice partners go to for advice?
Course Prerequisites
Participants should have taken a course in social network analysis, so be familiar with such terms as mutuality, transitivity, degree, centrality, n-cliques, structural equivalence, geodesic, row-stochastic. Participants should also have taken a basic course in (logistic) regression analysis. A basic course on networks structure is advisable.
Reading
Carrington, P.J., J. Scott, and S. Wasserman. 2005.Models and Methods in Social Network Analysis.Cambridge University Press.
Scott, J. 1992. Social Network Analysis. Sage.
Wasserman, S., and Faust, K. 1996. Social Network Analysis. Cambridge University Press.