Temporal Two-Mode Network Analysis
Temporal Two-Mode (Bipartite) Network Analysis · Also known as: temporal bipartite network analysis, dynamic two-mode network analysis, time-varying bipartite network analysis, longitudinal affiliation network analysis
Temporal two-mode network analysis tracks relationships between two distinct classes of nodes — such as authors and publications, or actors and events — across multiple time points. By combining bipartite structure with longitudinal observation, it reveals how affiliation patterns, collaborations, and community memberships form, evolve, and dissolve over time.
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 temporal two-mode network analysis when your data are longitudinal affiliation records linking two distinct entity classes — such as author–article, person–organisation, or firm–board memberships — and you need to understand how the structure of those affiliations changes over time. It is especially suited to bibliometrics, collaboration studies, interlocking directorates, and cultural participation research. Avoid it when the two node types are not genuinely distinct, when the observation period is too short or has only one time point (use standard two-mode analysis instead), or when sample sizes per slice are too small to yield stable network metrics.
Strengths & limitations
- Captures structural dynamics that a single aggregated network masks, such as the emergence or dissolution of clusters.
- Preserves the bipartite nature of the data without forcing artificial one-mode projections at the outset.
- Enables event-history and panel models that link network change to external covariates.
- Compatible with well-established projection and centrality measures adapted for bipartite graphs.
- Applicable across diverse domains: bibliometrics, organisational studies, ecology, and cultural sociology.
- Temporal slicing introduces researcher discretion: window width affects which dynamics are visible.
- Small slice samples reduce the stability and interpretability of centrality and community-detection outputs.
- Projection to one-mode networks discards information and can distort tie strength.
- Computational cost and complexity grow substantially with the number of nodes, edges, and time slices.
Frequently asked
What is the difference between two-mode and one-mode networks?
In a two-mode (bipartite) network, edges only connect nodes of different types — for example, people to organisations. In a one-mode network, edges connect nodes of the same type. Two-mode networks can be projected onto one mode (linking people who share organisations), but the projection loses information about the original affiliation structure.
How should I choose the width of temporal windows?
Window width depends on the process you are studying. For annual publication cycles, yearly windows are natural. For fast-moving online platforms, weekly or monthly windows may be needed. Sensitivity analysis — repeating key metrics with narrower and wider windows — helps verify that your conclusions are not an artifact of the chosen granularity.
Can I use community detection on a temporal bipartite network?
Yes, but with care. Community detection on bipartite networks requires algorithms that respect the two-mode structure, such as bipartite modularity maximisation (Barber 2007) or its temporal extensions. Applying standard one-mode algorithms to a projected network can produce misleading communities.
What software supports temporal two-mode network analysis?
Python's NetworkX and iGraph support bipartite graph operations and temporal slicing. The R packages igraph and bipartite offer dedicated two-mode functions. For longitudinal analysis, tnet (R) provides weighted bipartite centrality measures specifically designed for temporal data.
Is it valid to project before computing centrality across time?
Projecting first and then computing centrality for each temporal slice is common but introduces distortion because projection collapses shared-membership information. Where possible, compute bipartite centrality directly on each temporal slice and project only for specific analytical purposes, reporting both results.
Sources
- 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 ↗
- Latapy, M., Magnien, C., & Del Vecchio, N. (2008). Basic notions for the analysis of large two-mode networks. Social Networks, 30(1), 31–48. DOI: 10.1016/j.socnet.2007.04.006 ↗
How to cite this page
ScholarGate. (2026, June 3). Temporal Two-Mode (Bipartite) Network Analysis. ScholarGate. https://scholargate.app/en/network-analysis/temporal-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.
- Modularity AnalysisNetwork analysis↔ compare
- Social Network AnalysisNetwork analysis↔ compare
- Temporal Community DetectionNetwork analysis↔ compare
- Temporal Network AnalysisNetwork analysis↔ compare
- Two-mode Network AnalysisNetwork analysis↔ compare