Event Date
(Individual) Learning with GenAI Feedback:
A longitudinal Study on Students’ Engagement
Large Language Models (LLM) such as ChatGPT can provide formative and personalized feedback to improve text quality and learning success. Despite these advantages, it remains unclear to what extent L2 learners engage with and incorporate GenAI based feedback in the revision process, and which individual attitudes and external factors lead to feedback uptake. Therefore, the focus is to analyze the impact of GenAI-assisted feedback on students’ engagement and text quality in university German as a foreign language courses (CEFR, B2/ C1). In this presentation the focus is on the following research question: To what extent does ChatGPT-based written corrective feedback shape the writing development in terms of students’ perceptions? This longitudinal study adopts a sequential explanatory mixed-methods research design (QUAN + QUAL) to answer the research question. The data is used to carry out a hierarchical cluster analysis to identify latent learner profiles. The qualitative component of the study comprises the analysis of open-ended questions in reflection sheets as well as semi-structured interviews to get a holistic view of students’ engagement. The results show the learning paths of latent learner profiles and may be used to support the individual learning process in an era of pedagogical transformation.
Short Bio:
Stanislav Katanneck is currently a PhD student at the Université Catholique de Louvain (UCLouvain, Belgium) and works as a Language Education Consultant at the Goethe-Institut Houston, where he coordinates the “Deutsch Lehren Lernen” (DLL) program and professional development and continuing education initiatives across the United States and North America. His current research interests include (AI-supported) feedback, writing pedagogy, multilingualism, languages for specific and academic purposes, as well as the professional development and continuing education of teachers.