The Evolution of Networks and the Resilience of Interorganizational Relationships after Disaster

Abstract

This study uses social capital and evolutionary theory to examine organizational resilience in terms of interorganizational networks of disaster-struck organizations following Hurricane Katrina. Approaching post-disaster organizational resilience using social network analysis highlights the way pre-disaster relationships and networking patterns play a vital role in post-disaster rebuilding. Data support the idea of structural inertia, suggesting that disaster is an event that further strengthens pre-existing networks and is not a time when organizations might benefit from forging new networks in seeking support and fueling survival. Implications suggest that the social capital accrued through long-standing partnerships and efficient pre-disaster networking through building communities of practice significantly impact post-disaster resilience.

Keywords:

• Evolutionary Theory
• Organizational Resilience
• Social Capital
• Disaster
• Social Networks
• Hurricane Katrina
• Networked Forms of Organizing

Acknowledgments

This research was supported by a grant to the first author from the National Science Foundation, Award BCS-0554959. An earlier version of this paper was presented at the Sunbelt annual meetings of the International Network for Network Analysis, June, 2010, Riva, Italy.

Notes

1. Follow-up interviews were conducted with a subsample of participants and are not part of this study.

2. A total of 64 unique organizations were part of the initial sample. Eight of the original interviews could not be used in this study because of recording problems (n=4), the informant requested to not be recorded (n=3), or the substance of the interview revealed that the organization never reopened and was thus substantially different than the others in the dataset (n=1).

3. Contact the first author for the interview guide at [email protected]

4. The advantage of ATLAS.ti is that each instance of text that reflects codes can be labeled with all of the appropriate and relevant codes from the entire codebook. Quotes from interviews often revealed multiple theoretical themes and concepts simultaneously. The disadvantage is that a quantitative measure of intercoder reliability cannot be accurately calculated. The coders, however, noted the consistency with which they agreed about identifying episodes that needed coding (episodes could be a single phrase or as much as a paragraph or two) and that their code selections (from the codebook, Table 1) were consistently similar, too. Because of the extensive code list from which they could choose, they did regularly have minor differences about which codes each coder applied to the same episode. In all circumstances, coders were compelled to be sure their disagreements were resolved in a rigorous and thoughtful manner. No “standoffs” ever needed resolving by a third coder.

5. A chi-square test with Yates continuity correction revealed that the percentage of organizations that were resource-vulnerable did not vary by resilience, χ² (1, N=52)=1.44, p=0.23.

6. Tables reporting these results are available by contacting the first author at [email protected].