Testing Coleman’s Social-Norm Enforcement Mechanism: Evidence from Wikipedia
Executive Summary —
Harvard Business School professor Mikolaj Jan Piskorski and doctoral candidate Andreea Gorbatai look to the editing process on Wikipedia to test and validate the well-accepted (but little-verified) theory of sociologist James Coleman that social norm violations decline as network density increases. Support for Coleman's mechanism would alert us to the importance of punishments for norm violations and rewards for such punishments, and thus help us to design social systems in which norms are observed.Key concepts include:
- Coleman argued that high-density networks provide an opportunity structure within which third parties can compensate norm enforcers for the expense of chastising norm violators. Such payments encourage actors to punish those who violate norms, which in turn reduce the incidence of norm violation.
- Despite ubiquitous citations of Coleman's explanation, little empirical work has tested it convincingly.
- The researchers identified the improper use of the revert command by Wikipedia contributors-by which users can quickly knock out text they don't agree with and revert it back to a prior state-as a norm violation.
- The research found substantial support for the theory, suggesting that increasing network density to elicit norm compliance is justified.
- On Wikipedia, norm violations, punishments for such violations, and rewards for those who punish violators are all highly visible. Replicating these conditions in the design of a social system is critical; otherwise, norm violations will remain undetected and therefore unpunished.
Since Durkheim, sociologists have believed that dense network structures lead to fewer norm violations. Coleman (1990) proposed one mechanism generating this relationship and argued that dense networks provide an opportunity structure to reward those who punish norm violators, leading to more frequent punishment and in turn fewer norm violations. Despite ubiquitous scholarly references to Coleman's theory, little empirical work has directly tested it in large-scale natural settings with longitudinal data. We undertake such a test using records of norm violations during the editing process on Wikipedia, the largest user-generated on-line encyclopedia. These data allow us to track all three elements required to test Coleman's mechanism: norm violations, punishments for such violations, and rewards for those who punish violations. The results are broadly consistent with Coleman's mechanism.