<![CDATA[Limited evidence exists on how platform-specific AI feedback functions in secondary EFL writing assessment—especially regarding its connections to learners’ motivation and self-confidence. This mixed-methods study used a one-group pre–post design with 35 Indonesian secondary EFL students during regular classes, integrating Ginger Writer. Motivation items were adapted from Schmidt & Watanabe, and self-confidence items from Bandura; both scales used 5-point Likert responses and showed good internal consistency. Semi-structured interviews, focus group discussions, and non-participant observations explored students’ experiences with AI feedback during drafting and revision. Quantitatively, motivation increased significantly (pre: M=20.9, post: M=22.1; t(34)=−2.32, p=.026, Cohen’s d₍z₎≈0.39), while self-confidence rose modestly but not significantly (pre: M=20.7, post: M=21.5; t(34)=−1.80, p=.081, d₍z₎≈0.30). Qualitative data showed students describing immediate, non-judgmental feedback that supported iterative revision, error noticing, and sustained effort, along with limitations such as connectivity issues and the need for teacher mediation to interpret suggestions. Triangulation suggests that quick, actionable feedback is linked to increased motivation during revision cycles, whereas confidence may require more extended exposure and scaffolding. The study indicates that schools can feasibly incorporate platform-specific AI feedback into formative writing assessment when combined with teacher guidance and reliable access; policy should support teacher professional development and infrastructure to enable equitable implementation.]]>
Copyrights © 2026
