Network Theory and Analysis
What it is
Network theory and analysis treats social life as a web of nodes (people, groups, or organizations) joined by ties (relationships such as friendship, advice, or message exchange). Rather than asking what individuals are like, it asks how they are connected. Social network analysis is the toolkit that maps and measures these structures, turning the pattern of relationships itself into the object of study and explanation.
The core idea
The central claim is that structure shapes outcomes. Where you sit in a network, how central you are, whose ties you bridge, and how densely your contacts know one another can matter more than your personal attributes. Information, ideas, and influence travel along ties, so the shape of the web governs who learns what, who holds power, and how quickly a message or innovation spreads through a community.
How it is used
Researchers gather relational data, who talks to, trusts, or follows whom, and represent it as a graph or matrix. They then compute measures such as centrality (a node's prominence), density (how interconnected a group is), and brokerage (who bridges otherwise separate clusters). These metrics let scholars test how communication structures predict performance, diffusion, cohesion, or influence across teams, organizations, and online platforms.
In practice
In a company, two departments may rarely exchange information directly. A single employee who knows people in both becomes a broker: she occupies a structural hole and gains influence simply by bridging the gap, regardless of her rank. Mapping the advice network reveals her hidden importance, something a traditional org chart, which shows formal authority rather than actual communication, would entirely miss.
Key studies & evidence
Jacob Moreno's sociometry, introduced in Who Shall Survive? (1934), pioneered diagramming social ties and is widely treated as the origin of the network perspective. Mark Granovetter's The Strength of Weak Ties (1973) showed empirically that people more often hear about jobs through acquaintances than close friends, because weak ties bridge otherwise disconnected groups and carry novel information. Stanley Wasserman and Katherine Faust's Social Network Analysis: Methods and Applications (1994) consolidated the formal methods. Within communication, Peter Monge and Noshir Contractor's Theories of Communication Networks (2003) advanced a multitheoretical, multilevel framework linking generative mechanisms such as homophily, proximity, and exchange to the emergence of communication structures.
Critiques & limitations
Network analysis is strong on structure but often thin on content and meaning: a tie records that two people communicate, not what the relationship means to them or what is said. Boundary specification is a persistent problem, since deciding who belongs in a network can shape the findings. Static snapshots can miss how ties form and dissolve over time, though longitudinal and stochastic actor-oriented models increasingly address this. Critics also note a risk of structural determinism, treating position as destiny and underplaying agency, culture, and individual interpretation. Finally, relational data are demanding to collect, and missing ties can distort centrality and brokerage measures more severely than missing attribute data distort conventional surveys.
Applications
Network analysis is a workhorse of organizational communication, used to map advice, trust, and knowledge-sharing structures, locate informal leaders and brokers, and diagnose silos that formal charts conceal. In communication and information technology, it underpins social-media analytics: tracing how posts cascade, identifying influential accounts, and detecting communities and echo chambers on platforms. For AURA Lab work, the same logic applies to mediated and streaming environments, where viewer-streamer-chat interactions and follower graphs form measurable networks, and to social VR, where co-presence and proximity ties among avatars can be mapped to study how community and influence emerge in shared virtual spaces.