Social Network Analysis

Source Document: Nilufar Matin, Richard Taylor, John Forrester , Lydia Pedoth, Belinda Davis, Hugh Deeming, Maureen Fordham UoN. 31st May 2015. Mapping of social networks as a measure of social resilience of agents. Deliverable 4.2

Summary
The capacity of social network maps as a multi-purpose heuristic device is very useful – indeed necessary – if we want to explore ideas of community resilience and planning in the face of natural disasters. As White (1945) put it, if “floods are ‘acts of God’, flood losses are largely acts of man” and therefore being able to present a formalized, structured understanding of social aspects (“the social”) is critical to understanding, communicating, and providing truly integrative research on what community resilience to multiple hazards actually means in practice. Resilience cannot be left to hydrologists and physical planners alone, it must include the social. Our social network maps provide such a model of the links between significant individuals involved in key stages of the disaster planning, response and recovery phases. Further, they do so in a structured manner which allows this necessary social data to be communicated across disciplinary divides and also to be located within the context of a wider conceptual framework that links (social) knowledge networks and (institutional) decision-making structures. As a result, better decisions may be made.


There is a growing literature that suggests that social networks play a critical role in resilience to disasters (Aldrich 2012, Beilin et al. 2013, Hawkins and Maurer 2010, López-Marrero and Tschakert 2011, Tobin et al. 2014). Social networks can help, particularly in assessing how the network topology (structure, i.e., defined as specific patterns of connections between network actors, called ‘nodes’ in network terminology) and dynamics (processes, i.e., interactions among the nodes over time and space) influence the nature and level of resilience in a community. Further they can help identify barriers or structural holes in effective communication among actors; highlight aspects of power imbalances; and, provide a multi-actor multi-scalar view of interrelationships within and among communities.

The qualities that make communities resilient can be seen as emanating from unique abilities – inherent or learned – that a community embodies (Barrios 2014). Social network mapping allows us to capture that embodiment. Resilience to natural disasters was at one stage largely left to be a domain of hydrologists, seismologists, geologists, volcanologists, engineers and physical planners. Recent scholarship has challenged this notion and asserts that practices and structures of political and economic relationships, as manifested through social networks, are critical to resilience study. Capturing the structure of social relationships, and mapping them in time and space, contributes to our understanding of how community resilience emerges (cf. Barrios 2014: 330).

Structured maps are necessary for understanding and comparing relationships across the levels of governance and across scales of community. Traditional social science methods have on occasions fallen into the practical fieldwork-generated trap of focusing on one scale of community alone. This is unhelpful for understanding what community resilience is, or how it can be facilitated, for the emergent property of community resilience is certainly the produce of cross-scale social relationships that allow communities at local levels to become resilient (or not). The creation of community resilience can be seen as a social capital embodied in the quality of the relationships between people: and this relationship between people can be measured by network mapping.



Further, while traditional social science data-gathering methodologies, such as surveys and interviews, can be used for understanding resilience, only a structured approach to collecting and mapping social network data allows us to model social relationships. This formulation allows us to both measure and depict, and lends itself to attaining a level of social network analysis that can be used to feedback into not only our understanding of social capital or resilience, but also into policy formulation. As a research tool, maps have proved very useful at structuring the knowledge of a range of significant actors and re-presenting that knowledge in a way that is quickly and relatively easily usable and understandable by other actors in other positions in space and time. Social network mapping (henceforth SNM) thus can help (a) stakeholders to locate their place within a wider network, and (b) this network can be communicated and explained to others.

Unlike many other more ‘inductive’ social science methods, SNM is relatively front-loaded in terms of the attention that must be given to research design. To attain the precision required for SNM, it is often suggested that the research problem must be clearly identified at the beginning, before any data collection is done and the SNM approach is applied (Beilin et al. 2013, Tobin et al. 2014). However, examples are beginning to surface where from a rich variety of qualitative data, such as narratives or archival data, social networks can emerge (Emmel and Clark 2009, Edwards, G. 2010). In emBRACE we have employed both the approaches, as can be seen from the two emBRACE case studies that used SNM from each end, i.e. one gathering purposive quantitative data from surveys followed by participant interviews; and the other retrospectively culling network data from in-depth qualitative interview transcripts. The pros and cons of both of these approaches are discussed in the full report.


Ċ
Maureen Fordham,
Sep 23, 2015, 5:48 AM