ABSTRACT
There is a more or less standard problem that arises in the use of formal models in the study of social phenomena. Massarik and Ratoosh (1965, p. l7) have called it the "seductiveness of mathematics" and described it in these terms:
In this chapter a way around the problem will be suggested. A paradigm for a class of stochastic models will be introduced that requires a continual interplay between model and data. Use of this paradigm leads to the development of models that are specifically designed to guide the analysis of structural data. Because data are an integral part of model building in this case, the models are always tied directly to behavioral observations. Thus, what will be suggested is a procedure that draws upon the power of formal methods and, at the same time, requires that any formalisms developed are constantly kept in direct touch with their empirical roots.
