What's wrong with the diffusion of innovation theory?

Abstract

This paper examines the usefulness of the diffusion of innovation research in developing theoretical accounts of the adoption of complex and networked IT solutions. We contrast six conjectures underlying DOl research with field data obtained from the study of the diffusion of EDI. Our analysis shows that DOl based analyses miss some important facets in the diffusion of complex technologies. We suggest that complex IT solutions should be understood as socially constructed and learning intensive artifacts, which can be adopted for varying reasons within volatile diffusion arenas. Therefore DOl researchers should carefully recognize the complex, networked, and learning intensive features of technology; understand the role of institutional regimes, focus on process features (including histories) and key players in the diffusion arena, develop multi-layered theories that factor out mappings between different layers and locales, use multiple perspectives including political models, institutional models and theories of team behavior, and apply varying time scales while crafting accounts of what happened and why. In general the paper calls for a need to develop DOl theories at the site by using multiple levels of analysis. © 2001 by Springer Science+ Business Media New York.