Social Network Centrality Metrics

Social network centrality metrics are measures, such as degree, betweenness, closeness, and eigenvector centrality, that quantify how important or well-positioned each node is within a network.

사회연결망 중심성 지표는 연결중심성, 매개중심성, 근접중심성, 고유벡터중심성처럼 네트워크 안에서 각 노드가 얼마나 중요하거나 유리한 위치에 있는지를 수량화하는 측도이다.

How do I find the most important or influential actors in a network?

네트워크에서 가장 중요하거나 영향력 있는 행위자는 어떻게 찾나요?

Centrality metrics rank nodes by different notions of importance: degree counts direct ties, betweenness counts how often a node lies on shortest paths between others, closeness reflects distance to all other nodes, and eigenvector centrality rewards ties to well-connected nodes. They fit relational data such as friendship, collaboration, citation, or communication networks given as edge lists or adjacency matrices.

When to use Social Network Centrality Metrics

• An edge list or adjacency matrix of ties between people, organizations, or documents
• Identifying brokers, hubs, or key players in a network
• Using node centrality as a variable in a later regression

When to use something else

• The goal is to find groups of densely connected nodes: use community detection such as the Louvain method
• The goal is to test hypotheses about tie formation: use exponential random graph models

Assumptions and requirements

• Network boundary and tie definition are clearly specified
• Directed versus undirected and weighted versus unweighted ties are handled consistently
• Missing nodes or ties can bias centrality scores
• Closeness requires a connected network or a harmonic variant

What the analysis reports

• Degree, betweenness, closeness, and eigenvector centrality per node
• Ranking of top nodes on each measure
• Network-level statistics such as density and centralization
• Network graph with node size scaled by centrality

How to report it

[Node] had the highest betweenness centrality ([value]), indicating a brokerage position, while network density was [value].

Related methods

• Community Detection
• PageRank
• Exponential Random Graph Model (ERGM)

Run Social Network Centrality Metrics on Quorant