Co-Constructing AI Boundaries Research Methods Justification
Your approach is rigorous because it combines established qualitative methods (case study, micro-interactional analysis, triangulation) with your newly adapted theoretical framework (the 5-part M&M model). The inclusion of students as co-investigators adds a valuable layer of participatory rigor, which is well-regarded in qualitative and educational research.
🚀 Justification Assessment
| Approach Detail | Assessment | Connection to Rigor |
|---|---|---|
| Qualitative case study approach | Excellent. Appropriate for an in-depth, context-dependent investigation of a complex phenomenon (AI-mediated literacy practice). The provided article by Naeem & Thomas (2025) explicitly supports this strategy for studying complex issues with AI integration. | Aligns with Yin's (2014) and Stake's (1995) goal of rich, contextualized analysis. |
| Micro-interactional analysis of NotebookLM sessions | Strong. This method is the perfect fit for your 5-part framework, as it allows you to zoom in on Prompts, Outputs, and Integration at the moment the student is making a decision. | Provides the fine-grained data necessary to capture boundary-setting and moment-to-moment cognitive moves. |
| Comparative cross-case coding using the 5-part framework | Excellent. This explicitly turns the M&M pedagogical model into an analytic tool (your theoretical contribution). It provides structure and focus to your coding process, enhancing analytic rigor (Collins & Stockton, 2018). | Ensures systematic analysis across your data set and provides structure for interpreting complex events (Baxter & Chua, 2003). |
| Attention to evidence of boundary-setting (restriction, correction, refusal, modification) | Crucial. This is your primary analytic goal, focusing on agency and epistemic stance. These actions (restriction, refusal, etc.) are the visible enactments of the Mollick principle that the student must remain the "human in the loop." | Grounds abstract concepts (like boundary-work) in observable data behaviors, increasing validity and trustworthiness (Yin, 2014). |
| Triangulation with final projects and reflections | Required for rigor. Triangulation validates the findings from your primary data (micro-interactions) against secondary data (the final work and the metacognitive data). This specifically captures the Integration and Reflection stages of your framework. | Enhances the credibility and depth of your findings by converging evidence from multiple sources (Yin, 2018; Tellis, 1997). |
| Students participated as co-investigators, not subjects. | High Impact. This signals adherence to ethical and reflexive considerations (Creswell, 2013) and aligns with participatory research models, enriching the analysis by valuing the students' local knowledge and interpretations. | Addresses potential researcher bias and promotes inclusivity, strengthening the philosophical underpinnings of your qualitative study (Naeem & Thomas, 2025). |