Date of Original Version



Working Paper

Rights Management

All Rights Reserved

Abstract or Description

In this study, I examine the effects of corruption on patterns of relations in a well-known case of organizational crime. By contrasting corrupt and non-corrupt projects within an organization, I am able to examine how fraudulent endeavors alter the way individuals mobilize to accomplish a goal. This study examines longitudinal data and couples qualitative coding techniques with social network analysis to understand the effects of corruption on social structure. In contrast to non-corrupt projects, corrupt networks have lower connectivity, fewer reciprocal relations, and communicate less frequently. These different patterns also hold for between- and within-subject studies. For individuals, corrupt communications, as compared to non-corrupt communications, are less frequent, less likely to be reciprocated, and have reduced transitivity, meaning that message recipients are not as likely to share a communication link. This study highlights the role of content in understanding the emergent properties of communication networks.