{"id":2307,"date":"2025-10-07T18:59:45","date_gmt":"2025-10-07T16:59:45","guid":{"rendered":"https:\/\/sites.uef.fi\/learning-analytics\/?page_id=2307"},"modified":"2026-02-25T23:09:39","modified_gmt":"2026-02-25T21:09:39","slug":"tna-lak-workshop-2026","status":"publish","type":"page","link":"https:\/\/sites.uef.fi\/learning-analytics\/tna-lak-workshop-2026\/","title":{"rendered":"Transition Network Analysis Workshop"},"content":{"rendered":"\n

A Hands-on Tutorial with a Focus on Modeling Human-AI Interaction<\/strong><\/h2>\n\n\n\n

co-located with <\/strong>LAK 2026, Bergen, Norway<\/strong><\/h3>\n\n\n
\n \n
\n \"Mohammed\"\n <\/div>\n

Mohammed Saqr<\/p>\n

\n <\/svg> <\/div>\n <\/a>\n \n \n
\n \"Kamila\"\n <\/div>\n

Kamila Misiejuk<\/p>\n

\n <\/svg> <\/div>\n <\/a>\n \n \n
\n \"Volkan\n <\/div>\n

Volkan Y\u00fccepur<\/p>\n

\n <\/svg> <\/div>\n <\/a>\n \n \n
\n \"Sonsoles\"\n <\/div>\n

Sonsoles L\u00f3pez-Pernas<\/p>\n

\n <\/svg> <\/div>\n <\/a>\n \n <\/div>\n\n\n\n

Our workshop at LAK introduces Transition Network Analysis (TNA)<\/a>, a novel network analysis method that captures the relationships and dynamics of the learning process. The workshop offers attendees the opportunity to understand, learn, practice, and discuss several case studies using real-life data and hands-on experience. The workshop is an excellent opportunity to discuss the opportunities of applying TNA to one’s own data and get their questions answered by the creators of the methods. While the workshop will cover the broad applicability of TNA, the featured case studies will focus on human-AI interaction<\/a>\u2014an area of growing importance. This is due to the several new methods that TNA introduces specifically to address this area of significance.<\/p>\n\n\n\n

\"\"<\/figure>\n\n\n\n

What is TNA?<\/strong><\/h3>\n\n\n\n

TNA is a novel framework for modeling the learning process through a rich toolkit of rigorous, statistically validated methods. It provides an accessible yet powerful environment for researchers to perform sophisticated analyses\u2014without requiring coding expertise. TNA supports the computation of metrics at the graph, node, and edge levels, and facilitates the identification of recurring structures such as dyads, triads, communities, and clusters. TNA integrates statistical validation techniques \u2014including bootstrapping, permutation testing, and case-dropping\u2014 to assess the robustness and replicability of findings. This allows researchers to statistically validate each edge in the network with both p-values and effect sizes \u2014an unprecedented capability in process modeling. Furthermore, TNA enables comparison across subgroups and the explanation of observed patterns with edge-level statistical significance. By combining network analysis with statistical rigor, TNA supports theory development and hypothesis testing grounded in empirical evidence.<\/p>\n\n\n\n

The workshop\u2019<\/strong>s learning objectives<\/strong> include helping participants develop a solid understanding of theoretical foundations and methodological affordances of TNA and its key variants: Frequency-based TNA, Attention Network Analysis, and Co-occurrence TNA. Attendees will learn how to choose appropriate data and research questions for TNA, perform necessary data preprocessing, and apply the method using both point-and-click platforms where no code or programming experience is required (tna-web<\/a> and JTNA<\/a>) as well as code-based tools (t<\/em>na R package<\/a>).<\/p>\n\n\n\n

Key skills in the tutorial include<\/strong><\/h3>\n\n\n\n