Abstract For better or worse, vast amounts of the research in the social sciences is done by researchers working independently of each other. Moreover, until quite recently, most of the data gathered has been of the correlational variety, relating two variables at a time. Consequently, those who try to make theoretical sense out of the vast amount of data often give up this task as hopeless. One way of alleviating this situation is through a propositional inventory: by bringing into one volume a large number of propositions that cover similar grounds. The late 1950s and the 1960s saw an upsurge in such inventories and this trend has continued into the 1970s. 1 Yet even these inventories do not condense and organize the data sufficiently to permit one to develop theories from them. In this paper we shall describe and illustrate a procedure which permits one to construct simple theories from propositional inventories.