Enhancing Methodological Integrity with GenAI: A Multi-case Study of Experiential Learning using Sequential Augmented Analysis

Abstract

The rapid expansion of Generative Artificial Intelligence (GenAI) in higher education presents a critical pedagogical challenge for research training: how to integrate these tools without undermining methodological integrity. In qualitative research, unstructured GenAI use may encourage overreliance, superficiality, and unreflexive judgment among novice researchers. Despite growing debate, limited empirical evidence shows how GenAI can be deliberately designed to strengthen rigor in undergraduate qualitative data analysis. This study proposes and analyze the Sequencial Augmented Analysis a structured instructional model that embeds the use of chatbots within Human-Centered AI in Education and Experiential Learning frameworks. Using an exploratory multiple-case design with final-year pre-service teachers, we examined how GenAI-enhanced investigator triangulation and guided reflexivity support methodological integrity. Findings indicate that introducing chatbots after manual analysis stimulated collective reconsideration of decisions, systematic returns to original data, and clearer justification of methodological choices. It also surfaced personal biases, methodological assumptions, and ethical concerns regarding authorship and disclosure. Rather than replacing human judgment, GenAI functioned as a catalyst for dialogue and critique. The study offers a replicable pedagogical design for integrating GenAI into qualitative research courses while reinforcing academic integrity.