Organizational Complexity and Innovation: Developing and Testing Multiple Contingency Models - Management Science Vol. 42, No. 5.

Current research in organizational innovation is extensive, yet, because of limitations in scope, most studies are not adequately encompassing. These studies typically relate organizational variables to innovation and control at most for the effect of one contingency factor. Because innovation depends upon a complex host of factors, such theories have limited predictive application. This study intends to develop and test theories that explain the variation in the organizational complexity-innovation relationship in greater detail. The study considers two major indicators of organizational complexity—structural complexity and organizational size. Hypotheses are proposed on the effects of 14 contingency factors on the relationships between structural complexity and innovation and organizational size and innovation. The contingency factors include environmental uncertainty, organizational size, industrial sectors, types of innovation, and stages of innovation adoption. Using a meta-analytic procedure for multivariate analysis, the hypotheses are then tested with data from published studies in organizational innovation during the last three decades. The effects of four methods variables—operational definitions of innovation, structural complexity and size, and similarity of data sources—are controlled for in testing the hypotheses. This process results in two powerful and encompassing models: (1) the association between structural complexity and innovation depends upon operational definition of complexity, environmental uncertainty, use of manufacturing organizations, use of service organizations, focus on technical innovations, focus on product innovations, and focus on implementation of innovation; and (2) the association between organizational size and innovation depends upon operational definition of size, environmental uncertainty, use of service organizations, use of for-profit organizations, focus on technical innovations, and focus on product innovations. These models suggest avenues for further theory development and research, which we discuss.