Predicting and Understanding Organization Structure.

Size is examined as a predictor of organization structure with data from a British sample of business organizations, supplemented by findings from British labor unions, engineering firms, and the Aston sample of varied work organizations.' Where possible, comparison is also made with the results of an American study. While the broad outlines of formal organization structures are predictable with a high degree of confidence from a knowledge of organization size, a comparison of size-structure regressions across different industries suggests that other variables must also be taken into account. A solution, more satisfying both theoretically and statistically, emerges when complexity is distinguished from other aspects of structure which more directly constitute the framework of bureaucratic control-namely, the degree of formalization and decentralization. It is then found that while size, with technology, location and environmental variables, predicts complexity, the degree of complexity itself has a more direct relationship with formalization than does size. Size, however, remains the major predictor of decentralization. It is concluded that, in the organizations studied, complexity cannot be satisfactorily predicted or fully understood without reference to the economics of scale, but that it is neither theoretically convincing nor statistically demonstrable that size in itself is the major determinant of formalization.