An Examination of “Push-Pull” Theory Applied to Process Innovation in Knowledge Work

Abstract

Because many organizations have not been successful in introducing new task and managerial methods into the workplace, considerable attention has been directed toward developing a more complete understanding of organizational innovation. Three separate literatures (organizational science, engineering/R&D management, and MS/OR/MIS implementation) have, in fact, been concerned with innovation and technology diffusion; however, surprisingly little integration among the three has occurred. This paper reports on a study which examined a key tenet from each of these literatures in an effort to construct a robust model of innovative behavior. Specifically, the study utilized survey data in examining the validity of “push-pull” theory (i.e., that innovation is most likely to occur when a need and a means to resolve that need are simultaneously recognized) as well as the importance of top management attitude toward an innovation and of organizational receptivity toward change. The research context involved the diffusion of six modern software practices into 47 software development groups. While the model's independent variables explained a rather large amount of the variance in the use of these modern software practices, “push-pull” theory was not validated. A number of explanations are offered for the apparent failure of “push-pull” theory. Top management attitude and organizational receptivity toward change, however, were generally found to influence organizational innovation. As hypothesized, significant differences emerged in the factors influencing administrative and technical innovations with organizational receptivity toward change important only for the technical innovations. This suggests that organizational processes facilitating innovation should vary depending on the nature of the innovation involved.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 523 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.