Social Network Analysis Metrics

At WE&B we have been implementing Social Network Analysis (SNA) in the environmental sector over the past 5 years. We have a?scientific article published on the subject matter and various practical implementation examples from across different industries. We have brought greater insight to the reasons behind the connections between organisations. Insights that have provided networks with the critical information on who has the power, influence, and networking ability within and between these connections. However, what do all these metrics really mean??

SNA provides a measurable approach to analyse social relations and patterns of interaction among actors in a social network. A social network is characterised by underlying data which defines various features of a particular network. To visualise a social network in a concise manner, several measures – or metrics – are drawn upon to numerically define these features, as well as other influential nodes, in quantitative terms. Let's take a closer look at the more commonly used metrics to analyse social networks and what each of these metrics mean.?

• Node?is the name given to an individual, group, or organisation in a network map.
• Connection?is the name given to the lines (grey on the maps below) that connect one node to another.?

Degree Centrality?is the simplest measure of node connectivity. It assigns an importance score based simply on the number of links held by each node; that is, how many direct, ‘one-hop’ connections each node has to others in the network. For this reason, it can be useful to look at in-degree (the number of inbound connections) and out-degree (the number of outbound connections) as distinct measures to find either interconnected individuals, individuals who hold the most information, or individuals who can quickly connect with the wider network.

Our involvement in the?REWAISE project , for example, will help to create a new “smart water ecosystem” that involves stakeholders to embrace the true value of water through decentralised water services and decision-making. Thus, we are using SNA to identify local and regional stakeholders that are relevant to the project and, in particular, the?degree centrality?to help understand the network in terms of organisation connections and information flows.

Betweenness Centrality?quantifies the number of times a node acts as a bridge along the shortest path between other nodes. Nodes with high betweenness centrality are important to the information flow and cohesiveness of the network – they are considered to be central to the network due to their role in the flow of information in the network. Betweenness is particularly useful for analysing communication dynamics, though it should be used with care. A high betweenness count could indicate that someone holds authority over disparate clusters in a network or simply that they are on the periphery of both clusters.?

In the?HOUSEFUL ?project together we are seeking to enhance the systemic shift in the housing sector by providing the best strategy to move towards a circular economy, while also facilitating the decision-making process among stakeholders in the construction and housing sector. To this end, WE&B is using SNA analytics to map all relevant stakeholders in the entire value chain of the European Housing sector and, subsequently, we have applied the betweenness centrality to visualise the power, influence, and networking ability of each stakeholder.?

Closeness Centrality?calculates the shortest paths between all nodes, then assigns each node a score based on its sum of shortest paths; that is, it scores each node based on their ‘closeness’ to all other nodes in the network. Closeness centrality can therefore help find the individuals who are best placed to influence the entire network most quickly.

In the context of HOUSEFUL, one of the project main goals is to develop and demonstrate an integrated systemic service composed of 11 circular solutions co-created by stakeholders within the current housing value chain. It was thus vital to identify the good ‘broadcasters’ stakeholders in order to co-create circular solutions.?

Eigenvector Centrality?is a more sophisticated version of degree centrality. It not only measures a node’s influence based on the number of connections it has to other nodes in the network, but it also takes into account how well connected a node is and how many links their connections have through the network. In other words, eigenvector centrality depends on the number of incident links as well as on the quality of those links.

The??Run4Life project , “Recovery and Utilization of Nutrients 4 Low Impact Fertiliser”, proposed a radical new concept for wastewater treatment and nutrient recovery. The project was built around four demo-sites in Europe to advocate the institutional, legal, and social acceptance of its recovery technologies, as well as to categorise the stakeholders’ proximity to and involvement with the project. Therefore, WE&B undertook a SNA to effectively identify who are the relevant stakeholders, and, by way of eigenvector centrality, we could easily uncover how these stakeholders interacted between one another, what their influence and interest levels were in terms of the Run4Life technology systems.

PageRank?is a variant of eigenvector centrality, and, therefore, it also assigns nodes a score based on their connections and their connections’ connections. The difference is that PageRank takes link direction and weight into consideration, which is helpful for understanding citations and authority. Links can only pass influence in one direction and pass different amounts of influence, and, for this reason, this metric uncovers nodes whose influence extends beyond their direct connections into the wider network.

AfriAlliance is another project which had the main objective for African and European stakeholders to work together in the areas of water innovation, research, policy, and capacity development in order to prepare Africa for future Climate Change challenges. The SNA, that WE&B performed within this project has been helpful to give a visual overview of the AfriAlliance stakeholders and their connections. The Stakeholder Map is currently made up of approximately 623 connected elements that are either simple, combined, or temporary liaisons. With this many organisations and associations, we used PageRank to identify whose influence extends beyond their direct connections into the wider network.

If you want to know more about the SNA, or perhaps the ways in which we can provide you with this analysis and improved insights within your own organisational/departmental networks and your project networks, please visit our website on https://weandb.org or get in touch with us on?[email protected] .