Key points are not available for this paper at this time.
An attempt is made to explain an anomalous finding in an earlier survey by the authors (ibid, vol. EM-27, p.98-102, 1980) of the adoption of process innovations in the domestic footwear industry. Hypotheses that complexity would be positively related, and centralization and formalization negatively related, to the number of innovations adopted were confirmed by a correlation analysis but not by a zero-order correlation analysis. It is shown that complexity was acting as a suppressor variable, preventing the hypothesized relationships between formalization and adoption and between centralization and adoption from appearing in the zero-order analysis. Then the question is raised of why complexity has the power to override the effects of centralization and formalization. Case studies indicate several characteristics of decisions that reduce the hypothesized negative effects of high levels of centralization and formalization upon adoption. The case studies also suggest an interaction between firm structure and innovation characteristics: high levels of centralization and formalization reduce the adoption of revolutionary innovations but not of evolutionary innovations. This inference is tested and confirmed with data from the survey.
Cohn et al. (Thu,) studied this question.