Working on the edge: Examining covariates in multi-organizational networks following the September 11th attacks on the World Trade Center.

摘要

In the past ten years, the United States has experienced disasters that caused greater loss of life and property damage than the terrorist attacks of September 11, 2001. However, 9/11 led to large-scale changes in laws, policies, and government institutions, as well as historical changes to emergency management policies and procedures. This dissertation represents one of the most detailed, in-depth efforts to contribute to the understanding not only of this event, but of organizational behavior in disasters more generally. It focuses on the organizations, tasks, and activities that emerged immediately following the WTC attacks in lower Manhattan. To explain the observed relationships among organizations, this dissertation uses social network analysis to understand the factors that affect the interaction of organizations in the immediate aftermath of a disaster event. The relational data used in this analysis was collected and coded from more than 750 hours of field work, 2,249 newspaper articles, and 235 field documents to create a total set 6,661 interorganizational interactions among 727 organizations involved in 42 response tasks and activities. These organizations represent a variety of organization types (public, private, nonprofit, and collective) operating across a range of operational scales (local to international levels). This analysis applies a series of exponential random graph models with an inhomogenous Bernouilli family to examine the effects of exogenous covariates (type, scale, and level of involvement) in each of the forty-two task networks, as well as the aggregate response network. In the results, the structure of ties within networks and the factors that affect tie formation vary over the forty-two task groups reaffirming the diverse and complex range of interorganizational relations that characterize large-scale disaster response. Significant findings related to public-private and public-nonprofit partnerships show how pre-disaster interdependencies contribute to the formation of post-disaster emergent ties. These results show that patterns of assortative and disassortative mixing vary across response tasks and activities. The findings of this dissertation offer initial steps to understanding the distinctive features of individual disasters while offering an understanding of commonalities across disasters and determine characteristics that best serve as indicators of emergent social structure.