Modeling Homophily in Exponential-Family Random Graph Models for Bipartite Networks

Abstract

Homophily, the tendency of individuals who are alike to form ties with one another, is an important concept in the study of social networks. Yet accounting for homophily effects is complicated in the context of bipartite networks where ties connect individuals not with one another but rather with a separate set of nodes, which might also be individuals but which are often an entirely different type of objects. As a result, much work on the effect of homophily in a bipartite network proceeds by first eliminating the bipartite structure, collapsing a two-mode network to a one-mode network and thereby ignoring potentially meaningful structure in the data. We introduce a set of methods to model homophily on bipartite networks without losing information in this way, then we demonstrate that these methods allow for substantively interesting findings in management science not possible using standard techniques. These methods are implemented in the widely-used ergm package for R.