Association for Public Policy Analysis & Management

There are important questions as to why disaster management systems are unable to take timely and coordinated action in response to disruptive events. According to Elinor Ostrom (2005, p. 48-49), decision-makers are often constrained by conditions of uncertainty, which represent the indeterminacy that occurs when “institutional arrangements leave open wide avenues for choice, and each individual’s outcome is dependent upon the actions taken by others.” When operating under conditions of uncertainty, the decision-makers within response organizations may know the type of action that they should take, but information deficiencies undermine their ability predict which of their actions will enable them to obtain their desired outcomes.

 While conventional wisdom suggests that policy-makers can improve the capacity of organizational systems through resilience (United Nations 2007), the manner in which resilience is facilitated and maintained is not fully understood (Manyena 2006). Theory does suggest, however, that organizational systems that possess the capacity to adapt to conditions of uncertainty are resilient (Comfort 1997). A system’s capacity to adapt, or modify the structure of its activities, is related to many factors, including its social-technical characteristics and the availability of information (Haase 2010). This paper will investigate the presence of structural change within the response system that operated in Indonesia after the Great Sumatran Earthquake and Tsunami. Indonesia was selected as a case study due to the scope and size, as well as the international character of the tsunami disaster. This study addresses three research questions: (1) what was the organizational composition of the response system; (2) what were the structural characteristics of the response system; and (3) to what extent did the structure of the response system evolve?

 These questions will be answered though an analysis of newspaper articles and daily situation reports published between 26 December 2004 and 17 January 2005. These data were coded, first, by identifying the name, source of funding, and level of jurisdiction of the organizations that participated in the response system. We also plotted longitudinally, by date, the rate that organizations became active, as well as the number and type of interactions undertaken by each organization. The second coding step identified the organizational interactions, which were transformed into a series of twenty-four relational matrices and examined using network analysis software. We then computed two common network level measures to reveal and evaluate the structure of the response network: density and degree centralization (Wasserman and Faust 1994). We calculated these statistics for each of the twenty-four relational matrices included the period under analysis, and then plotted the results longitudinally, by date, to evaluate whether we could detect structural change within the response system. These findings indicate that the structure of the response system changed. In terms of theory, these findings reveal the presence of resilience, which is a mechanism that enables administrative systems to adapt and maintain a continuity of operations in conditions of uncertainty. In terms of practice, these findings will help disaster managers recognize that steps can be taken to strengthen the capacity of disaster response systems.