Directed Two-Mode Network Analysis
Directed Two-Mode (Bipartite) Network Analysis · Also known as: directed bipartite network analysis, asymmetric affiliation network analysis, directed actor-event network analysis, directed two-mode graph analysis
Directed two-mode network analysis studies bipartite graphs in which nodes belong to two distinct sets — such as actors and events, authors and papers, or firms and markets — and edges carry a direction, capturing asymmetric relationships like citation, referral, or endorsement. Combining the duality of two-mode structure with directed tie semantics reveals flow patterns and influence asymmetries that undirected or single-mode analyses would miss.
Read the full method
Sign in with a free account to read this section.
Method map
The neighbourhood of related methods — select a node to explore.
When to use it
Use directed two-mode network analysis when your data involves two conceptually distinct node types and asymmetric ties between them: citation networks linking papers to references, user-content interaction networks where users post to forums, supply-chain graphs where firms source from suppliers, or actor-event networks where attendance implies a directional relationship. It is appropriate when each mode contains at least several dozen nodes and when directionality is theoretically meaningful. Do not use it when ties are genuinely symmetric (mutual co-membership), when you have only one type of node, or when mode membership is ambiguous — forcing two-mode structure on one-mode data introduces artefacts.
Strengths & limitations
- Captures asymmetric flow and influence between two distinct node types that single-mode analysis conflates.
- HITS scoring naturally decomposes the network into hub and authority roles across the two modes.
- Preserves the duality of affiliation: both modes are analysed together rather than one being collapsed away.
- Applicable to a wide range of empirical data types — citation, firm-market, user-content, and biological interaction networks.
- Directional community detection reveals gatekeeping and brokerage roles invisible to undirected bipartite models.
- Projection onto one mode discards information and inflates tie density, potentially distorting results.
- Standard centrality measures require adaptation for bipartite structure and may lack efficient implementations in common software.
- Collecting reliable directional tie data for both modes is often labour-intensive and error-prone.
- Interpretation requires domain knowledge to correctly assign direction and to distinguish substantively meaningful from artefactual asymmetries.
Frequently asked
How is directed two-mode analysis different from ordinary directed SNA?
Ordinary directed SNA operates on a single set of nodes with directed ties among them. Directed two-mode analysis maintains a strict partition into two node sets and only allows ties between modes, never within a mode. This bipartite constraint enables HITS-style hub-authority decomposition and requires distinct centrality measures.
Should I project the bipartite graph to a one-mode graph before analysis?
Projection is sometimes useful for applying standard SNA tools, but it loses information and inflates density. Whenever possible, analyse the bipartite graph directly using methods designed for two-mode structures. If you must project, document the projection rule and acknowledge the information loss.
Which centrality measure is most appropriate for directed two-mode graphs?
HITS (hub and authority scores) is the most natural choice because it was designed for directed bipartite graphs — hubs in one mode point to authorities in the other. In-degree and out-degree can also be computed per mode. Betweenness and closeness require adaptation to bipartite paths and should be used carefully.
What sample size is needed?
There is no universal threshold, but sparse graphs with fewer than about 20 nodes per mode produce unstable centrality estimates and unreliable community detection. Graphs with at least 50 nodes per mode and a meaningful number of directed ties per node yield more interpretable results.
Can directed two-mode analysis handle weighted ties?
Yes. Weighted directed bipartite graphs represent tie strength such as citation counts or transaction volume. Weighted versions of HITS and degree centrality are straightforward; weighted bipartite community detection is available in packages such as igraph and NetworkX.
Sources
- Wasserman, S. & Faust, K. (1994). Social Network Analysis: Methods and Applications (Ch. 8). Cambridge University Press. ISBN: 978-0-521-38707-1
- Borgatti, S. P. & Everett, M. G. (1997). Network analysis of 2-mode data. Social Networks, 19(3), 243-269. DOI: 10.1016/S0378-8733(96)00301-2 ↗
How to cite this page
ScholarGate. (2026, June 3). Directed Two-Mode (Bipartite) Network Analysis. ScholarGate. https://scholargate.app/en/network-analysis/directed-two-mode-network-analysis
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
- Directed Community DetectionNetwork analysis↔ compare
- Directed Modularity AnalysisNetwork analysis↔ compare
- Directed Social Network AnalysisNetwork analysis↔ compare
- Knowledge Graph AnalysisNetwork analysis↔ compare
- Multiplex Network AnalysisNetwork analysis↔ compare
- Two-mode Network AnalysisNetwork analysis↔ compare