A graphical chain model derived from a model selection strategy for the sociologists graduates study

This paper objects to the arising problems due to fitting graphical
chain models to multidimensional data sets. This multivariate
statistical tool is used to cope with complex research questions
concerning not only direct, but also indirect associations between
the variables of interest. Due to this high complexity sensible
strategies for fitting such models are required. Here, a
data--driven selection strategy is discussed. Its application is
illustrated for an empirical data example in detail.