Knowledge seeking and provision in work teams across multiple knowledge areas: A test and extension of Transactive Memory Theory in the network approach.

摘要

This study aims to examine Transactive Memory theory both in the descriptive and normative aspects, by conducting analyses in three stages. In the first stage, important information seeking and provision patterns were identified in 12 pre-existing work teams. While some of these patterns agree with the propositions of transactive memory theory, others are not mentioned or are in opposition to transactive memory propositions. Subsequently, the impacts of these transaction patterns on group knowledge development and group performance were investigated. Most patterns of directional information seeking advocated by TM theory were found to improve group knowledge level and centralization, while the corresponding patterns of directional information provision did not show much influence. Finally, none of the knowledge transaction patterns were found to have significant impacts on group performance in terms of goal attainment. On the one hand, this finding serves as a caution that group knowledge development and group performance are distinct concepts and should be measured and analyzed separately. Increasing the amount of or the centralization of group knowledge does not automatically improve group performance. On the other hand, these results are limited to the patterns included in the model. More information transaction patterns should be examined, and multiple performance measures should be employed to capture the impacts of information transaction on different aspects of group performance.;As a serious attempt to analyze information transaction patterns across groups and knowledge areas in a coherent way, this study conducted multilevel network analyses, and attempted to go beyond simple descriptions and pursue statistical inference of some level. Contributions and limitations are listed. Directions for future research and practical implications are suggested at the end.