Validation problems in games and simulations with special reference to models of international politics

To what extent are simulations and games valid representations of behavioral systems? Although this frequently asked question is vital, it concerns only one aspect of the larger issue of operating model validation. This article describes how the purpose of the experimenter, the type of validity criteria, and the use or nonuse of human participants each influences validation. Many of the illustrations are drawn from games and simulations of international politics, but the observations are intended to be applicable to similar activities underway throughout the behavioral sciences.