Network models for social selection processes

We present network models for social selection processes, based on the p* c lass of models. Social selection occurs when individuals form social relati onships on the basis of certain characteristics they possess. Similarity is a common hypothesis for selection processes, but one that is usually frame d dyadically. Structural balance approaches move beyond dyadic conceptualiz ations and require more sophisticated modeling. The two-block chain graph a pproach of p* social influence models is adapted to allow individual attrib ute variables to be predictors of network ties. Using a range of dependence assumptions, we present a hierarchy of increasingly complex selection mode ls, including models for continuous attribute measures, which in their simp lest form may be assumed to be linear. The models have scope, however, for more complex functional formulations so that more specific hypotheses may b e investigated by postulating a particular functional form. Our empirical e xamples illustrate how dyadic selection may be transmuted into structural e ffects, and how the absence of dyadic selection may still mask a subtle hig her order selection effect as individuals ''position" themselves within a w ider social environment. In conclusion, we discuss the links between social influence and social selection models. (C) 2001 Elsevier Science B.V. All rights reserved.