A simple model of sociometric choice is presented, assuming dyad independence, a common assumption in social network models. Possible simplifications by choice or edge independence are also discussed. The study concentrates on structural sociometric variables such as the number of reciprocal choices. Data are from a survey of children in sixth grade where the children were asked to select their best friends in class. The different models are evaluated using goodness-of-fit statistics. The large number of networks available makes it possible to find and evaluate empirical distributions of the test statistics.

 Key words: dyad independence, choice independence, edge independence. 

Close this Window

Last update: 1997-12-16 / KH