Theoretical Decision Rules for Selecting Research Designs

Since nation-states are complex, they require special methodological rules for sampling. The suggestion is made that societies should be selected on the basis of their score on some independent variable and studied over long time periods so as to maximize information gain and to minimize cost. Then the choice of societies to represent alternative explanations and finally to represent particular kinds of control should be made. The problem of identifying the appropriate frequency of measurement is discussed via a number of general classes of sociological and political variables. If these rules of purposive sampling are followed, one approximates an experiment that will build on present theoretical knowledge.