Organizational memory: Integrative framework and empirical tests of its performance benefits in different task environments.

摘要

This dissertation introduces two constructs that help develop the concept of organizational memory. A demand-side construct of social retrieval consists of social and non-social sources, differentiated by the human interaction requirement. A supply-side construct of risk-based storage consists of high-risk and low-risk archives, differentiated by the probability of information loss linked to employee turnover. Examples include people and reference manuals.;Empirically, this dissertation focuses on the demand-side construct and tests the effects of access to social and non-social sources of organizational memory on task performance in different environments. The tests involve a 2x2x4 factor experiment where student members of a Finance Club solving business plan problems over a continuous 11-hour period. The experiment utilizes published scales, strong incentives, stratified random assignment, Latin Square block design, and repeated measurement procedures to account for possible measurement errors, such as individual differences, learning-by-doing, self-selection, boredom, and exhaustion.;A generalized linear mixed model was used to analyze data of 246 students answering questions about four business plans with each business plan consisting of 30 questions (n=29,520). Results provide support for eight of ten hypotheses. First, without requisite knowledge, the effect of access to social sources alone on task performance is greater than the effect of access to non-social sources alone. Second, without requisite knowledge, the effect of access to non-social sources reduces with access to social sources. Third, knowledge interacts with social sources and non-social sources; hence, with requisite knowledge, the effect of access to social sources alone on task performance reduces more than the effect of access to non-social sources alone. Fourth, the effect of requisite knowledge on task performance increases when tasks become more complex. Fifth, the effect of access to non-social sources on task performance does not change when tasks become more complex (hypothesis not supported). Sixth, the effect of access to non-social sources on task performance increases when tasks become more complex. Seventh, complexity interacts with social sources more than with non-social sources; hence, the effect of access to social sources alone on task performance is greater than the effect of access to non-social sources alone when tasks become more complex. Eighth, the effect of perceived specialization of information sources on task performance does not change with requisite knowledge (hypothesis not supported). Ninth, the effect of perceived credibility of information sources on task performance reduces with requisite knowledge. Finally, the effect of perceived coordination of information sources on task performance reduces with requisite knowledge.