Learning-by-Participating: The Dynamics of Information Aggregation in Organizations
45 Pages Posted: 25 Jul 2019
Date Written: July 24, 2019
In the Carnegie School tradition, work on organizational decision-making is underpinned by two streams of research: the process of learning-by-doing, where individuals learn from feedback on their own choices; and the process of information aggregation, where multiple individuals’ beliefs are aggregated to organization-level decisions via a decision-making structure (e.g., plurality voting). These streams have evolved somewhat independently, and in fact, research on information aggregation often assumes that individuals do not learn (i.e., they hold stable beliefs). We bridge these streams with the observation that, in organizations, individuals often receive feedback not on their own choices, but rather, on the choice made by the organization. We refer to this as “learning-by-participating.” The existence of learning-by-participating implies that we must recognize that alternative decision-making structures vary not only with respect to (a) their effectiveness in aggregating individual beliefs, but also with respect to (b) their effectiveness in shaping individual beliefs via learning. Our core insight is that these forces are often in conflict. We find that the outcome of this conflict turns on the way in which alternative decision-making structures draw on and shape the diverse knowledge of the organization’s participants, and in particular of contrarians—individuals whose choices differ from that of the organization. Implications include the observation that decision-making structures that are effective when individuals’ beliefs are stable may be much less effective when individuals can learn; and, in contrast to the wisdom-of-crowds logic, larger organizations (crowds) may not be substantially more effective than smaller ones.
Suggested Citation: Suggested Citation