Peers are of great importance to children and young people. The social networks in which they are embedded can influence their individual behaviour, attitudes and performance. Statistical network methods explicitly address the interdependencies between actors within a network. This approach allows the investigation of a wealth of research questions when social processes are the focus.

Tom Snijders presented his explanations using the SAOM model framework. This is a statistical model that can be used to analyse panel data in a network context, including individual attributes. Snijders explained that the model is based on the insight that over time, each actor in the network is influenced in its behaviour by the network and adapts its behaviour in the network. The network therefore influences and changes individual behaviour over time, just as individual behaviour influences and changes the network. The two processes run simultaneously in mutual interdependence.

Closing, Snijders discussed some examples of the application of SOAM to situations in the educational context. He presented studies on social network relationships and students' choice of academic subjects. Snijders also did not fail to mention limitations (e.g., the difficulty of specifying network boundaries) and open problems (e.g., deriving measures of effect size) of the approach and invited the audience to discuss them.

Tom Snijders is professor emeritus of methodology and statistics at the University of Groningen and professor emeritus of statistics in the social sciences at the University of Oxford, where he is also an emeritus fellow of Nuffield College and an associate member of the Department of Statistics.

Link [external] to Tom Snijders Website