TNA Bern Workshop Resources
A curated hub of R packages, Python ports, interactive apps, hands-on tutorials, and references for the Bern Transition Network Analysis workshop.
Bern · Workshop Resources
Transition Network Analysis (TNA) models sequences of events as weighted directed networks, making it possible to uncover the temporal dynamics, transitions, and relational patterns that drive behavioral, psychological, and educational processes. The tna R package brings the method to practice with four model families — transition (Markov), frequency-based, co-occurrence, and attention-weighted — together with bootstrap stability, permutation-based group comparisons, community detection, and rich visualisation. Nestimate wraps and extends this stack into a unified estimation framework that also covers psychological networks and panel data. This page collects the packages, interactive Shiny and jamovi apps, hands-on tutorials, Nestimate articles, blog posts, and the published papers that ground the method — so you can move from raw event logs to validated network models in one place.
tna R Package
The reference R implementation of Transition Network Analysis. In addition to the tna R package, you can also try our Shiny app — an interactive web application for exploring sequences, communities, centrality, and network visualisation — and our Jamovi plugin, which brings TNA into the jamovi point-and-click environment with built-in statistical validation.
Apps & tools
- Dynalitics app https://sonsoleslp.shinyapps.io/dynalytics/ Interactive Shiny app for dynamic and temporal learning analytics — explore sequences and transitions in the browser without writing R.
- Jamovi website https://www.jamovi.org/ Free, open-source statistical software — a friendly alternative to SPSS/SAS and the host environment for the JTNA plugin.
- Jamovi guide https://sonsoles.me/extra/JTNA.html Step-by-step tutorial for running Transition Network Analysis inside jamovi using the JTNA plugin — no coding required.
Tutorials
| Tutorials | Link |
|---|---|
| An Updated Comprehensive Tutorial on Transition Network Analysis (TNA) A modern, end-to-end walkthrough of rigorous behavioral process analysis with tna — the recommended starting point. | https://sonsoles.me/posts/tna-tutorial/ |
TNA Data Preparation: A Comprehensive Guide to prepare_data() How to turn raw event logs into analysable sequences using prepare_data(), which builds sessions from time gaps, ordering columns, or actor groupings. | https://sonsoles.me/posts/tna-data/ |
| TNA Group Analysis: Analysis and Comparison of Groups Estimate separate transition networks per group and compare them with permutation-based significance tests. | https://sonsoles.me/posts/tna-group/ |
| TNA Clustering: Discovering and Analysis of Clusters Discover latent subgroups in sequence data using mixture Markov clustering, then inspect each cluster’s distinct transition pattern. | https://sonsoles.me/posts/tna-clustering/ |
| TNA Model Comparison:TNA Model Comparison: A Comprehensive Guide to Network Comparison Compare TNA models — edges, centralities, and full networks — with bootstrap and permutation procedures for valid statistical inference. | https://sonsoles.me/posts/tna-compare/ |
| Full reference guide on tna functions Official function reference for tna v1.1.0 covering all four model families plus bootstrapping, stability, community detection, and visualisation. | https://sonsoles.me/tna/tna.html |
| Sequence Patterns, Outcomes, and Indices with codyna Hands-on tutorial for codyna — sequence-level indices, recurring patterns, and outcome associations to complement TNA. | https://sonsoles.me/posts/codyna-seq-tutorial/ |
Nestimate R Package
Unified network estimation, analysis, and validation for behavioral, psychological, and panel data.
https://saqr.me/Nestimate/A single interface for transition networks (transition, frequency, co-occurrence, attention-weighted), psychological networks, and panel-data networks — with bootstrap, simplicial-complex, and model-assessment tools layered on top.
Apps
- Co-occurrence app https://saqr.me/cooccureapp/ Interactive web app for building co-occurrence networks from raw data — auto-detects input format and computes similarity measures in the browser.
- cograph app https://saqr.me/cographapp Point-and-click companion to the cograph package — load an edge list and explore communities, centrality, and visualisations without code.
Full tutorials
-
Network Estimation & Visualization
End-to-end guide: estimate a transition network with Nestimate and visualise it with cograph. -
Higher-Order & Simplicial Complexes
Move beyond pairwise transitions to higher-order structure — from transition networks to topological analysis with simplicial complexes. - Validate transition networks with split-half reliability, case-dropping stability, and Markov-order adequacy checks.
Quick guides
- The basics: estimate and analyse a transition network in Nestimate.
- Visualise sequences as heatmaps, index plots, and distribution plots — with side-by-side group comparisons.
-
Clustering & Multi-Level Analysis
Discover sequence clusters and run multi-level analyses to compare networks across nested groups. - Markov-stability community detection — uncover communities that persist across multiple time scales.
- Worked example comparing early vs late human–AI interactions to detect shifts in sequence patterns.
Featured blog post
Human–AI Interaction: A Transition Network Analysis with cograph
https://saqr.me/blog/2026/human-ai-interaction-cograph/ Sequential analysis of 13,002 turns across 429 sessions of human–AI coding collaboration, mapping the transition structure of the dialogue with TNA and cograph.Related packages
Sibling R packages and the Python port that round out the TNA ecosystem.
- cograph · R Tools for the analysis, visualisation, and manipulation of dynamical, social, and complex networks — supports heterogeneous, multi-layer, and hierarchical networks with a simple syntax. Used throughout Nestimate’s visualisation pipeline.
- codyna · R Analysis of complex dynamic systems — the temporal unfolding of patterns, changes, and state transitions in behavioural data. Complements TNA with sequence-level pattern detection and outcome association.
- tnapy · Python Full Python port of the TNA framework with numerical equivalence to the R package. Supports eight model types (relative, frequency, co-occurrence, attention-weighted, n-gram, gap-weighted…), network plots, heatmaps, and centrality charts — installable with
pip.
References
-
Saqr, M., López-Pernas, S., Törmänen, T., Kaliisa, R., Misiejuk, K., & Tikka, S. (2025). Transition Network Analysis: A Novel Framework for Modeling, Visualizing, and Identifying the Temporal Patterns of Learners and Learning Processes. In Proceedings of the 15th International Learning Analytics and Knowledge Conference (LAK ’25) (pp. 351–361). ACM. https://doi.org/10.1145/3706468.3706513
The foundational TNA paper (LAK ’25) introducing the framework for modelling, visualising, and identifying temporal patterns in learning processes. -
Tikka, S., López-Pernas, S., & Saqr, M. (2025). tna: An R Package for Transition Network Analysis. Applied Psychological Measurement. https://doi.org/10.1177/01466216251348840
The official software paper for the tna R package — cite this when you use the package. -
Saqr, M., López-Pernas, S., & Tikka, S. (2025). Mapping Relational Dynamics with Transition Network Analysis: A Primer and Tutorial. In M. Saqr & S. López-Pernas (Eds.), Advanced Learning Analytics Methods: AI, Precision and Complexity. Springer. https://lamethods.org/book2/chapters/ch15-tna/ch15-tna.html
Book-chapter primer that models transitions between events as a weighted directed network to capture temporal dynamics and relational patterns in learning. -
Saqr, M., López-Pernas, S., & Tikka, S. (2025). Capturing The Breadth and Dynamics of the Temporal Processes with Frequency Transition Network Analysis: A Primer and Tutorial. In M. Saqr & S. López-Pernas (Eds.), Advanced Learning Analytics Methods: AI, Precision and Complexity. Springer. https://lamethods.org/book2/chapters/ch16-ftna/ch16-ftna.html
Companion chapter on Frequency Transition Network Analysis (FTNA) — models transition frequencies directly, without Markov assumptions. -
López-Pernas, S., Tikka, S., & Saqr, M. (2025). Mining Patterns and Clusters with Transition Network Analysis: A Heterogeneity Approach. In M. Saqr & S. López-Pernas (Eds.), Advanced Learning Analytics Methods: AI, Precision and Complexity. Springer. https://lamethods.org/book2/chapters/ch17-tna-clusters/ch17-tna-clusters.html
Heterogeneity-focused chapter: use mixture Markov models to identify subgroups with distinct transition patterns.