Distributing "raw" data among types or classes is a necessary and illuminating part of the process of research and discovery in any science, particularly in the early stages of the latter's development. But it produces fruitful results only if the types or classes make sense, which they will just to the extent that, inter alia, the variables we fix upon in defining them are the significant ones (for the purpose in view, of course), and that the classes (a) exhaust the phenomena under consideration, and (b) do not overlap.
One of the most elementary procedures used in dealing with the raw data of political conflict is that which, taking its departure from the notion of "party systems," seeks to assign each observed instance to one or another of three types: the "one-party system," the "two-party system," and the "multiple-party system." All party systems, it is assumed, belong as a matter of course to one of the three, so that one of the researcher's first tasks in studying the phenomena of party conflict in a given political situation is to find out with which one of the three types he is dealing. Until he has done this—so runs the tacit premise—he does not have his problem in manageable shape.