3
Social Network Analysis for Improved Disaster Preparedness and Intervention Planning

Chapter 2 summarized presentations and discussions of the first session of the workshop regarding case studies and real-world examples of social network analysis (SNA) application. This chapter summarizes Sessions 2 and 3 of the workshop (see Box 1-2 for workshop agenda). Session 2 was devoted to exploring the use of social networking and SNA for disasters and public health emergencies; it was focused on the empirical identification of social support networks, the measuring of networks and understanding communication among them, and on assessing opportunities for interventions to improve community disaster resilience. Workshop participants broke into two groups that met concurrently. Session 2a addressed the role of communication and ways to improve the functional, structural, and interactional linkages between networks to foster preparedness. Session 2b examined the opportunities and constraints in the application of SNA for planning interventions that improve community disaster resilience.

Session 3 was devoted to exploring the use of social networking and SNA for improvisational disaster response. Opportunities to use SNA to facilitate emergency preparedness and response were examined for two different venues: networks among and between organizations; and networks within local communities. As in Session 2, workshop participants were broken into two groups that met concurrently. Session 3a addressed how SNA could be used to make interactions among networks of organizations more productive and improvisational responses more flexible. Session 3b focused on how individuals and communities could be empowered and engaged to foster collective behavior while preparing for, responding to, and recovering from disasters.

As described in Chapter 1, questions for guiding discussion were prepared in advance for each session by the workshop planning committee. A member of the planning committee served as moderator for each session. Descriptions of each session and the guiding questions were provided to workshop participants prior to the workshop, and are provided in Appendix D of this document. Workshop participants regrouped for summary discussions at the end of each session. Discussions in these sessions were generally more abstract and conducted at a higher level than in Session 1. This level of conceptualization is reflected in the remainder of the summary.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 33
3 3 ________________________________________________________________________ Social Network Analysis for Improved Disaster Preparedness and Intervention Planning Chapter 2 summarized presentations and discussions of the first session of the workshop regarding case studies and real-world examples of social network analysis (SNA) application. This chapter summarizes Sessions 2 and 3 of the workshop (see Box 1-2 for workshop agenda). Session 2 was devoted to exploring the use of social networking and SNA for disasters and public health emergencies; it was focused on the empirical identification of social support networks, the measuring of networks and understanding communication among them, and on assessing opportunities for interventions to improve community disaster resilience. Workshop participants broke into two groups that met concurrently. Session 2a addressed the role of communication and ways to improve the functional, structural, and interactional linkages between networks to foster preparedness. Session 2b examined the opportunities and constraints in the application of SNA for planning interventions that improve community disaster resilience. Session 3 was devoted to exploring the use of social networking and SNA for improvisational disaster response. Opportunities to use SNA to facilitate emergency preparedness and response were examined for two different venues: networks among and between organizations; and networks within local communities. As in Session 2, workshop participants were broken into two groups that met concurrently. Session 3a addressed how SNA could be used to make interactions among networks of organizations more productive and improvisational responses more flexible. Session 3b focused on how individuals and communities could be empowered and engaged to foster collective behavior while preparing for, responding to, and recovering from disasters. As described in Chapter 1, questions for guiding discussion were prepared in advance for each session by the workshop planning committee. A member of the planning committee served as moderator for each session. Descriptions of each session and the guiding questions were provided to workshop participants prior to the workshop, and are provided in Appendix D of this document. Workshop participants regrouped for summary discussions at the end of each session. Discussions in these sessions were generally more abstract and conducted at a higher level than in Session 1. This level of conceptualization is reflected in the remainder of the summary. 33

OCR for page 33
34 APPLICATIONS OF SNA FOR BUILDING COMMUNITY DISASTER RESILIENCE SNA FOR IMPROVING COMMUNICATION Session 2a participants (see workshop agenda, Box 1-2) identified three inter-related long-term goals for improving communication for disaster preparedness and intervention: 1. The development of a networking planning tool, such as desktop computer software, to support homeland security. The tool could be designed to be used by an individual with at least some college education to define and model a given community. Such a tool would allow the user to identify community members, the resources available to the community, and the best pathways to disseminate information to the community to achieve a desired outcome; 2. A better understanding of networking theory. This includes processes by which individuals emerge as leaders within a network, and what happens to information disseminated across a network. With a better understanding of network theory, knowledge of various processes can be leveraged to support community resilience, create more effective support for communities, and to mobilize resources quickly when necessary; and 3. The means to conduct impact analysis (e.g., what happens in a network when specific information is disseminated) and scenario analysis (e.g., which of multiple scenarios will have the most desired impact). Tools are currently available to accomplish some of these objectives, and new technologies are rapidly evolving. However, many existing tools may be too academically oriented for practical application or are not packaged for the specific needs of the disaster management community. The most effective user interfaces for computer programs could be developed with the full cooperation of emergency management practitioners and an understanding of the practitioners’ needs. Workshop participants noted that to get the most out of networking tools, practitioners will need training in their use. Interpretation of network visualization graphs (Figure 2-1, for example) will need to be incorporated into the training of law enforcement and first responders. Training in the use of networking tools would also be essential at the community level to educate the public on topics such as how to send text messages, and how to communicate with family members during emergency events. The next three sections of this document discuss in greater detail the three objectives stated above. A Network Planning Tool for Practitioners Many workshop participants stated that a network planning and visualization tool for the emergency management practitioner cannot be designed to perfectly plan for every disaster or reach every individual in a community. However, such a tool could be invaluable in helping managers maximize connectivity within networks and building community disaster resilience. Building the tool could be accomplished in phases. Initial phases could allow basic SNA functions as described in Chapter 2. Emerging technologies, such as data mining

OCR for page 33
SNA FOR IMPROVED DISASTER PREPAREDNESS AND INTERVENTION PLANNING 35 and different analysis techniques, could be incorporated into future phases. The goal of the primary phase could be to provide planning support for the Department of Homeland Security (DHS), community leaders, and other community groups. Initial challenges will likely be indentifying the capabilities the tool should have, identifying aspects of networks that need to be targeted, getting data into the tool, and developing the capacity to maintain and update the data. Workshop participants stated that for planning purposes, the tools would give practitioners an understanding of a community’s adaptive capacities during different emergencies. Sufficient flexibility to accommodate dynamic networks and technologies would make the tool more useful. Workshop participants considered lessons learned from organization theory research. Optimizing a network could reduce the adaptability of that network; for example, optimizing a network to increase the speed with which goods are delivered during nonemergencies may involve choosing specific resources and transportation modes. Without backup plans and flexibility during times of crisis, people may be put at risk if the goods are unavailable from regular sources, or infrastructure fails. If the goal is to increase resiliency, it is essential that the network’s adaptive capacities—its abilities to function under stress—are encouraged to expand and remain flexible. Given this information, workshop participants discussed incorporating network optimization functions into planning and visualization tools only with proper caution so as not to limit flexibility. The Theory of Networking A better understanding of network theory can lead to better knowledge of communication processes. This kind of knowledge can be leveraged to support community disaster resistance, resilience, and recovery. The practitioner may be better informed on how networking tools such as Twitter and Facebook can be used to advantage. Understanding how networks and networking tools function is vital for practitioners, but it is also essential that practitioners understand how to create and distribute a robust message so that the correct message is sent to appropriate networking sites and media. Marketing experts may provide insight regarding how to create and distribute correctly received messages. Theory has not yet been developed on how to accomplish this. Participants indicated the importance of basic research on concepts related to trust and online networks. Community members have different levels of trust in data sources and in the digital networking tools through which data are transmitted. For example, many individuals may be distrustful of information coming from community leaders or from networking sites community leaders support. Understanding issues of trust can help practitioners more effectively use social networks to convey their messages. Understanding the processes by which individuals or groups of network members gain trust within their networks to become online opinion leaders is another area to be explored. Practitioners may be able to take advantage of the same processes and gain the trust of networks, or they may be able to predict who the opinion leaders will be and meaningfully engage with these potential leaders to encourage the spread of important information.

OCR for page 33
36 APPLICATIONS OF SNA FOR BUILDING COMMUNITY DISASTER RESILIENCE On the infrastructure side of the issue, the means of managing bandwidths required for social networking is challenging. Understanding the impact of the use of various cyber-enabled communication tools during a disaster situation is a new endeavor, and difficulties encountered when encouraging the public to use the communication channels designated by emergency managers is exacerbated during disaster situations. Theory building and data analyses in these areas would also be beneficial. Impact and Scenario Analyses The third of the long-term communication goals described earlier is the development of tools to conduct impact and scenario analyses. The tools could include those that (1) explore the best ways to disseminate negative news without making a bad situation worse (for example, what happens in a network if a specific warning is sent to a specific media outlet); and (2) choose the best of several options given a specific scenario (for example, what happens if electricity is restored first to one location rather than another). Tools that allow the user to understand the impact of information and actions were considered useful by workshop participants, but their use is largely dependent on an understanding of networking theory. Workshop participants noted that the need for tools for impact and scenario analyses is not unique to disaster management. Commercial tools are under development for marketing purposes, and the Department of Defense is developing tools for security purposes. Multiple private sector organizations are exploring tracking capabilities using text mining and text analysis techniques. Quickly and visually tracking the rapid changes that occur during an emergency is challenging, especially when changes are monitored over a variety of media. Challenges in the Use of SNA and Networks Workshop participants observed that quality data are required to conduct SNA analysis, and that such data are often unavailable. The lack of data and the capacity to manage large datasets are impediments to the use of SNA for real-time applications. In some cases, data exist to populate SNA tools, but there may be legal issues regarding the use of private information by public entities, or reluctance among jurisdictions and organizations to share data. The use of networking tools, as noted by workshop participants, is also somewhat incompatible with the DHS National Incident Management System (NIMS) guidelines for managing domestic incidents.1 Quality baseline data were also considered essential to effective SNA. The current state of a community has to be understood before effective interventions can be optimally designed and implemented, and their impacts measured. However, some workshop participants stated that the cost of collecting baseline data through interviews, fieldwork, and other means is prohibitive. 1 The DHS issued and revised the National Incident Management System (NIMS) to serve as a template to manage incidents of any size regardless of the cause, location, or complexity. See www.fema.gov/emergency/nims/index.shtm (accessed April 20, 2009).

OCR for page 33
SNA FOR IMPROVED DISASTER PREPAREDNESS AND INTERVENTION PLANNING 37 The cost of developing and maintaining new technologies is also high. Champions of networking technologies within communities and in Washington, D.C. (where they could interact with the federal government or appropriate interest groups) could be identified and asked to encourage the development and use of network planning and analysis tools. The most effective champions would be able to communicate the utility of networking and analysis tools to people such as first responders and those empowered within the community to overcome political obstacles. Partnering with groups that are already developing tools for impact and scenario analyses may be an effective means of advancing their development and use for emergency management purposes. Including emergency management practitioners in all stages of the conversation regarding the promotion of networking technologies would likely yield the most promising results. SNA FOR PLANNED INTERVENTIONS Participants of Session 2b (see workshop agenda, Box 1-2) discussed three main issues: 1. The ways that knowledge of social networks is currently used to support disaster management and build resilience; 2. Additional research needed to help improve disaster management and community resilience, namely, additional SNA of formal disaster networks, community-based networks, and the intersection between the two; and 3. Funding strategies to support the required research. Integration of Networking and SNA into Emergency Management Policy and Practice Workshop participants shared examples of how some practitioners and policy makers have at least a rudimentary understanding of social networks. Local communities in Southern California, for instance, have tapped into existing social networks to disseminate preparedness information. Lay health educators from the Latino community use their existing social networks to distribute information on earthquake preparedness and to share coping strategies across communities. American Red Cross staff are seeking to build organizational linkages with business partners, creating a more robust network of people able to staff emergency shelters. SNA is being incorporated to some degree in models to measure the effectiveness of epidemic disease containment measures in the case of pandemic influenza. Practitioners and researchers alike raised numerous questions regarding the integra- tion of social networking and SNA into management policy and practice. According to some workshop participants, the research literature already answers some of these ques- tions, but primary research is still needed on a variety of topics and scales related to dis- aster management agency responsibilities, community networks, and the interactions between them. For example, does being prepared for disaster within one network (e.g., where an individual is employed) make an individual more prepared in another network (e.g., at home or within a religious community)? The effectiveness of using networks to

OCR for page 33
38 APPLICATIONS OF SNA FOR BUILDING COMMUNITY DISASTER RESILIENCE promote resilience, disaster readiness, response, and recovery at different levels of soci- ety down to the household level could be studied. The identification and roles of faith- and community-based organizations within a community need to be explored. Compara- tive studies on the ways social networks affect levels of community resilience in diverse places such as Israel, the United Kingdom, and Japan would help build understanding of the network characteristics that are successful in this context. Several questions were raised by workshop participants regarding the role of organizational culture (for example, fire, police, emergency medical services, and public health departments) in building social networks, in the sharing of information in the disaster context, and in building resilience in general. Research on the ideal network among these groups and the role of technology in facilitating it could provide the guidance needed to create the robust and flexible networks that are sustainable during a disaster. Research on networks that cut across districts or local, state, and federal levels of government and disaster management goals could lead to study of how these networks could be improved. Emergency management practitioners would benefit from understanding the difference between information and resource networks, as well as the implications of those differences, so that they may effectively use networks to plan and implement interventions. Understanding how to identify and communicate with the correct people and organizations within different types of networks would also be useful. Research on the characteristics and behaviors of groups that are effective at organizing themselves for action around the hazards in their communities could inform practitioners about which community behaviors are advantageous and which are not. Similarly, research on how emergent groups use social networking technologies, for what purposes, and how these activities could be improved would also be important. A New Research Funding Strategy Several workshop participants observed that the research infrastructure does not support the longitudinal research necessary to understand and apply SNA for building community resilience. A substantial infusion of research support would be essential to develop and maintain both longitudinal and rapid-response research. Workshop participants discussed the idea of creating regional collaboratives of local universities, agencies, and businesses. These collaboratives could be funded with local, state, and federal resources, and could serve as repositories for regional baseline data. Their existence could encourage thorough baseline expertise on regional social networks and adaptive capacities, and they could be information resources for federal and local response agencies during times of crisis. The regions could be consistent with the 10 regions into which the United States is divided by FEMA. Both longitudinal and rapid- response research could be conducted within the framework of the collaboratives.

OCR for page 33
SNA FOR IMPROVED DISASTER PREPAREDNESS AND INTERVENTION PLANNING 39 SNA FOR ENHANCING IMPROVISATIONAL RESPONSE WITHIN NETWORKS OF ORGANIZATIONS In thinking about how organizations work together during a disaster, the participants of Session 3a (see workshop agenda, Box 1-2) considered improvisation an important aspect of disaster response. SNA has the potential to reveal new ways to coordinate or influence the convergence of people, resources, and information to improve improvisational response—those activities planned in immediate response to changed conditions or resources. Different organizations (e.g., police, public utilities) are responsible for different aspects of response, and all benefit from knowing what the others are doing. An organization or individual that responds in unexpected ways outside the organizational mission context can be like a musician playing a sour note; discord can be the result. Questions arise regarding how to involve individuals and groups in the response process; who is responsible for doing so; and how is information about their involvement shared among networks. The data essential for making these decisions are often unavailable. The next sections of this chapter summarize workshop discussion of how networks of networks may be used to foster improvisational response, communication, and resilience. Tools for Fostering Sustainable Ties SNA is a useful tool for understanding the nonlinear nature of many ties within networks and organizations. It may also help identify how flexibility could be built into networks and organizations to allow for effective planning given uncertain circumstances. Practitioners are most successful working with teams that continue to function sustainably during the unexpected. Of importance to the practitioner is how to create ties with those groups with interests relevant to a particular problem. For example, developing ties with animal rights groups may be beneficial when dealing with problems associated with the increased number of stray animals following a disaster. Better ties with a large number of organizations will likely yield greater network resiliency during times of disaster. A greater probability of successful and timelier response and recovery may also result. Baseline data help researchers understand the conditions necessary for building successful relationships to achieve desired outcomes. Because disasters are not contained within jurisdictional or geographical boundaries, building ties and brokering information across agencies and jurisdictions could prove effective. The knowledge base is increased and additional resilience is built into the combined networks. Baseline data and SNA may also help emergency practitioners determine the needed balance between efficiency and redundancy when developing relationships for resilient networks. Establishing redundancy in a network requires resources, but is essential in situations when a part of the network fails. Some workshop participants cautioned that redundancy that creates competiveness among network members should be avoided. Sustainable ties can be built with interfaces between government and community- based networks. Traditionally, communication during a disaster has been one-way, from emergency management to the public. Many workshop participants predicted that two-way

OCR for page 33
40 APPLICATIONS OF SNA FOR BUILDING COMMUNITY DISASTER RESILIENCE communication will be central to future disaster responses. The practitioner will be most effective by staying informed of the dynamic nature of relationships within a network and how these relationships may change day to day and in response to stress. This is essential to sustainable communication within the network. Technologies available today can already provide practitioners with many pieces of information that when combined, tell a story of what a network is doing and how it may be changing. Practitioners can use cell phones to monitor movement and receive status updates from the emergency management community or the public using networking tools, such as Twitter, to stay informed. SNA research conducted to determine how best to monitor constantly changing and emerging networks would aid practitioners. Avoiding infringement of privacy rights would be an aspect of this research. Chaos and Improvisation Workshop participants discussed the critical need for baseline research to understand how a community and its networks function normally and under stress. Some functioning may be chaotic and emergency managers could benefit from understanding when chaos does not need to be controlled. Some chaos may appear in the form of ad hoc groups that spontaneously offer assistance. Effectively working with these groups without a plan or knowledge of their capabilities depends on the ability of practitioners to direct volunteer response efforts appropriately. Impromptu enthusiasm can be harmful if not controlled. For example, the delivery of unneeded supplies could create added burden for emergency managers and distract them from more essential responsibilities. SNA may help determine where volunteer group efforts can be used. Research on how to foster the ability to rely more on improvisational response could be useful. According to workshop participants, determining who will direct these groups is as important as how to direct them. A census will never exist that lists all resources available during a disaster. Disaster management practitioners do not know what groups exist or who will ultimately be able to provide services once a disaster occurs. They may not have a full understanding of the critical dependencies for decision making (for example, who depends on whom and for what). Such gaps in knowledge limit the ability to improvise effectively. Unpredictable failure and re-emergence of networks add to the chaos. SNA research and the development of tools that help practitioners sort through this kind of chaos could be valuable. Networks that emerge or re-emerge following a disaster may be unstable or loosely constructed. Understanding how these emerging networks are organized is more complex the bigger the disaster. Understanding how mistakes may occur during the re-emergence of networks, and of how to correct them, is essential if practitioners are to use the networks effectively in disaster response. Networks of Networks The way SNA tools can be used to understand how networks function has already been discussed in this summary. SNA tools can be used beyond postdisaster case studies

OCR for page 33
SNA FOR IMPROVED DISASTER PREPAREDNESS AND INTERVENTION PLANNING 41 of individual networks. They could be used to understand how networks of networks emerge and evolve in a disaster situation, and may help identify the sustainable linkages among them. Given enough data, the commonalities and important characteristics that allowed these networks and linkages to survive can be discovered. However, accessing and using public data for the purpose of studying or constructing networks of networks is often challenging. Knowing what public data exist and getting access to them are challenging issues that are inadequately resolved by the Freedom of Information Act (FOIA),2 according to some workshop participants. Jurisdictions have independent protocols for recording data that may be incompatible with one another. As already mentioned, jurisdictions and agencies may be reluctant to share information. Some of the most useful data for analyzing and constructing networks could be the informal or confidential data representing personal communications between individuals and between different organizations. There are no means to obtain data from informal sources. The ways the fire, emergency management systems, police, and public health organizational cultures help or hinder the process of building networks of networks for disaster management and resilience is not well understood. Understanding the optimal role of networking technologies in creating sustainable ties among the organizations could enhance communication and collaboration during normal operations as well as during emergencies. Because disasters do not proceed according to a plan, emergency managers may be able to use SNA and related technologies to help networks of organizations coordinate improvisational responses. 2 The full text of the Freedom of Information Act (5 U.S.C. § 552, as amended by Public Law No. 104-231, 110 Stat. 3048) can be found at www.usdoj.gov/oip/foia_updates/Vol_XVII_4/page2.htm (accessed April 20, 2009).

OCR for page 33