A METHODOLOGICAL FRAMEWORK FOR CHATGPT TO ENHANCE ACADEMIC WRITING PROFICIENCY IN FIRST-YEAR UNIVERSITY EFL STUDENTS
Keywords:
ChatGPT, Generative AI, Academic Writing, EFL Pedagogy, Zone of Proximal Development, Scaffolding Framework, Formative Feedback, Prompt Engineering, Higher Education, Educational Technology.Abstract
This comprehensive study establishes a rigorous methodological framework for utilizing generative Artificial Intelligence (AI), specifically OpenAI's ChatGPT, to enhance the academic writing proficiency of first-year university students learning English as a Foreign Language (EFL). Moving away from traditional restrictive measures or unstructured AI utilization, this paper presents the 'AI-Scaffolded Academic Writing Framework' (AI-SAWF). Rooted in Lev Vygotsky's Zone of Proximal Development (ZPD) and social constructivism, the framework repositions ChatGPT as a dynamic, interactive pedagogical scaffold rather than a passive text generator. Employing a mixed-methods explanatory sequential design, the study analyzed 120 first-year undergraduate students over a full 16-week academic semester, divided equally into an Experimental Group (utilizing the AI-SAWF model) and a Control Group (relying on traditional peer and instructor review). Quantitative evaluations based on standardized IELTS Academic Writing rubrics showed statistically significant advancements (p < 0.05) for the experimental cohort across four critical dimensions: task achievement, coherence and cohesion, lexical resource, and grammatical accuracy. Qualitative analysis derived from thematic coding of focus-group interviews, prompt logs, and reflective teacher journals demonstrated a marked reduction in writing anxiety and a profound advancement in algorithmic prompt literacy and critical thinking. Ethical hazards, such as academic plagiarism, cognitive atrophy, and data hallucinations, are critically discussed, providing clear strategic countermeasures for contemporary higher education. The paper concludes with comprehensive curriculum reform proposals to foster responsible AI literacy.
References
1. Barrot, J. S. (2023). Using ChatGPT for academic writing: Opportunities, challenges, and future directions. Journal of English for Academic Purposes, 64, 101260.
2. Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative Research in Psychology, 3(2), 77-101.
3. Creswell, J. W., & Creswell, J. D. (2018). Research design: qualitative, quantitative, and mixed methods approaches (5th ed.). SAGE Publications.
4. Hyland, K. (2019). Second language writing (2nd ed.). Cambridge University Press.
5. Jeon, J., & Lee, S. (2023). Large language models in EFL education: A systematic review of ChatGPT integration. Computer Assisted Language Learning, 36(8), 1145-1169.
6. Kasneci, E., Sessler, K., Sternecker, S., van Duin, M., & Kasneci, G. (2023). ChatGPT for education: Scope for progress and potential dangers. Education and Information Technologies, 28(7), 8421-8449.
7. Krashen, S. (1985). The Input Hypothesis: Issues and Implications. Laredo Publishing Co.
8. Liu, M., & Mitchell, R. (2023). Formative feedback in the age of AI: Comparing ChatGPT and human tutors in EFL context. Language Learning & Technology, 27(2), 45-68.
9. MacArthur, C. A. (2023). Evaluation and revision of academic texts using generative artificial intelligence tools. Writing Research Quarterly, 15(1), 12-29.
10. Vygotsky, L. S. (1978). Mind in society: The development of higher psychological processes. Harvard University Press.
11. Warschauer, M., Tate, T., & Zheng, B. (2023). Digital scaffolding and AI in language education: A socio-constructivist perspective. Annual Review of Applied Linguistics, 43, 89-105.
12. Zhai, X. (2022). ChatGPT for education: A paradigm shifts in assessment and curriculum design. SSRN Electronic Journal, 4323926.
13. iversity Press.
