Pedagogical Judgment in Evaluating and Adapting Generative AI Outputs for Teaching English to Young Learners: A Pre-Service Teacher Case Study

Abstract

This study examines how pre-service English teachers evaluate and adapt generative artificial intelligence (GenAI) outputs for Teaching English to Young Learners (TEYL). Using a qualitative case study design, it involved 48 sixth-semester students enrolled in a Practicing of TEYL course at UIN Salatiga during the 2025/2026 academic year. Data were collected through classroom and peer-based microteaching observations, interviews with 10 students, and analysis of AI-assisted lesson plans and teaching materials. The data were examined thematically, with evidence from different sources compared to support the credibility of the findings. The findings show that students primarily used GenAI to generate initial ideas for lesson plans and activities, including stories, songs, games, role-play, and Total Physical Response (TPR). However, they found that AI-generated materials were often generic and required adjustment to address young learners’ developmental characteristics, language proficiency, interactional needs, and cultural contexts. Students revised these materials through lesson planning, peer-based microteaching, feedback, reflection, and subsequent revision. The study’s main contribution is a contextual account of AI-mediated pedagogical judgment, showing how pre-service teachers assess and adapt GenAI outputs in relation to the pedagogical demands of TEYL. As the teaching activities took place in peer-based microteaching rather than actual young-learner classrooms, the findings do not establish the effectiveness of the adapted materials with children. The study highlights the importance of preparing pre-service teachers to use GenAI critically while retaining responsibility for pedagogical decisions.