This study aimed to analyze students’ emotional intelligence development in STEM robotics learning through the Lesson Study framework integrated with Transcript-Based Lesson Analysis (TBLA), NVivo visualization, and Rasch Model analysis. This study employed a mixed-methods approach involving 23 junior high school students participating in robotics learning activities at SMPIT Nurul ‘Ilmi Jambi, Indonesia. The research was conducted in two Lesson Study cycles consisting of Plan–Do–See stages. Qualitative data were collected through classroom observations, audio-video recordings, and interaction transcripts analyzed using TBLA and NVivo software. Quantitative data were obtained through an emotional intelligence questionnaire and analyzed using the Rasch Model. The findings revealed significant improvements in students’ collaboration, communication, and emotional involvement from Cycle I to Cycle II. NVivo visualizations showed dominant emotional regulation expressions such as “wait,” “patient,” and “connect it first,” indicating the development of self-regulation, empathy, persistence, and teamwork during robotics troubleshooting activities. Rasch analysis further confirmed that students demonstrated a high level of emotional intelligence, with an average person measure of +1.19 logits and good model fit (Infit MNSQ = 1.03; Outfit MNSQ = 1.00). The study concludes that robotics learning within the Lesson Study framework effectively promotes students’ emotional intelligence through collaborative problem-solving and reflective interaction. The integration of TBLA, NVivo, and Rasch Model analysis provides a comprehensive framework for evaluating socio-emotional competencies in STEM education. The findings imply that STEM robotics learning should integrate emotional and collaborative dimensions as essential competencies for twenty-first century learners.
Affiliations come from OpenAlex; journal OAI-PMH metadata carries none. Authors without an OpenAlex ID are matched by name, which can merge or split people.