SocialNetworkAnalysen

SocialNetworkAnalysen is a methodological approach to studying social structures by mapping and measuring relationships among actors. The core idea is that the pattern of connections among individuals, groups, or organizations influences behavior, information flow, influence, and outcomes. In practice, networks are modeled as graphs: nodes (actors) and edges (ties). Edges may be directed or undirected and can carry weights indicating tie strength or frequency of interaction. Analyses distinguish egocentric networks (centered on a focal actor) from sociocentric networks (encompassing all actors in a defined boundary).

Common measures include centrality (degree, betweenness, closeness, eigenvector), which assess position and influence; cohesion and clustering; and community structure detection. The field also studies structural holes, brokerage roles, and network dynamics over time.

Data are collected from surveys, administrative records, or digital traces from emails, social media, or collaboration platforms. Methods must address missing data, measurement error, sampling bias, and privacy concerns, particularly with sensitive information. Temporal networks require longitudinal designs and dynamic models to capture changes.

Applications span organizational analysis (informal hierarchies, knowledge flows), epidemiology (contact patterns and disease spread), innovation and diffusion (adoption pathways), marketing (influence networks), and online platforms (friendship, follower, or interaction networks).

Software tools commonly used include Gephi, Pajek, UCINET, and Python libraries such as NetworkX. The approach draws on sociology, graph theory, statistics, and computer science, and remains subject to interpretational limits when inferring causality from observed associations.