Statistical modelling of the group structure of social networks

• This research evaluates the identification of latent classes in social networks.
• For the historical network studied, the model identification is consistent with the original study.
• A new Bayesian model comparison method is used to determine the number of latent classes.
• This is based on the posterior distributions of the competing model deviances and of their class membership probabilities.
• This finds that two classes are sufficient.

This research evaluates the identification of group structure in social networks through the latent class model and a new Bayesian model comparison method for the number of latent classes. The approach is applied to a well-known network of women in Natchez Mississippi. The latent class analysis reproduces the group structure of the women identified by the original sociologists.