Modeling Dependencies in International Relations Networks

Michael D. Ward Department of Political Science, University of Washington, Seattle, WA 98195 e-mail: isere{at}msn.com Despite the desire to focus on the interconnected nature ofpolitics and economics at the global scale, most empirical studiesin the field of international relations assume not only thatthe major actors are sovereign, but also that their relationshipsare portrayed in data that are modeled as independent phenomena.In contrast, this article illustrates the use of linear andbilinear random–effects models to represent statisticaldependencies that often characterize dyadic data such as internationalrelations. In particular, we show how to estimate models fordyadic data that simultaneously take into account: (a) regressorvariables, (b) correlation of actions having the same actor,(c) correlation of actions having the same target, (d) correlationof actions between a pair of actors (i.e., reciprocity of actions),and (e) third-order dependencies, such as transitivity, clustering,and balance. We apply this new approach to the political relationsamong a wide range of political actors in Central Asia overthe period 1989–1999, illustrating the presence and strengthof second- and third-order statistical dependencies in thesedata.