Enhancing Adolescent Narrative Writing through AI Intervention: A Pre-Post Quasi-Experimental Study of Sudowrite in Indonesian Secondary Education
Keywords:
AI-assisted writing, Narrative competence, EFL (English as a Foreign Language), Sudowrite, Indonesian secondary educationAbstract
This quasi-experimental study examined the potential contribution of Sudowrite, a generative artificial intelligence (AI) writing assistant, to the enhancement of adolescent narrative writing in Indonesian English as a Foreign Language (EFL) classrooms. Thirty-six Grade 11 students with CEFR A2–B1 proficiency from a private high school in Surabaya, Indonesia completed folklore-based narrative tasks ranging from 300 to 500 words before and after a six-week intervention. This intervention included phases of orientation, guided practice, independent application, and metacognitive reflection. The intervention employed Sudowrite to facilitate idea generation, plot development, vocabulary enhancement, and grammatical accuracy, with educators ensuring ethical implementation while maintaining student creativity. The narrative quality was evaluated through a validated five-criterion analytic rubric that assessed structure, idea development, characterization, language use, and mechanics, resulting in an inter-rater reliability exceeding 0.85. Paired-sample t-tests suggested statistically significant improvements in all criteria (p < .001), accompanied by large effect sizes (d = 0.82–1.34). Significant improvements were observed in idea development and language use, suggesting that AI-mediated feedback may help address linguistic limitations and support more advanced storytelling. Qualitative data revealed enhanced writing confidence while highlighting the necessity for explicit instruction to avoid stylistic uniformity. The single-group design restricts causal inferences and generalizability; however, the findings offer preliminary evidence that well-integrated AI tools may improve narrative competence while preserving cultural authenticity. This research supports a hybrid teaching method that integrates AI-driven micro-feedback with human facilitation to enhance technical skills and creative expression in secondary EFL settings. Recommendations involve ensuring equitable access to AI, providing comprehensive training for teachers, and conducting longitudinal multi-site research to investigate knowledge transfer, retention, and demographic variations within an Indonesian educational context.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 SEAQIL Journal of Language Education

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.




