@article{ART003195730},
author={Oh, Hyun-Joo and Hsieh, Shu-Fei},
title={An Analysis of ChatGPT-Generated Feedback on Argumentative Essays in Korean EFL Classrooms},
journal={Modern English Education},
issn={1598-0782},
year={2025},
volume={26},
pages={166-181},
doi={10.18095/meeso.2025.26.1.166}
TY - JOUR
AU - Oh, Hyun-Joo
AU - Hsieh, Shu-Fei
TI - An Analysis of ChatGPT-Generated Feedback on Argumentative Essays in Korean EFL Classrooms
JO - Modern English Education
PY - 2025
VL - 26
IS - null
PB - The Modern English Education Society
SP - 166
EP - 181
SN - 1598-0782
AB - In response to the increasing use of ChatGPT, this study aimed to analyze ChatGPTgenerated feedback on 126 argumentative essays written by university students in Korean English as a Foreign Language (EFL) classrooms, focusing on content, organization, language and grammar, and mechanics. Results revealed that ChatGPT assigned the highest score to content but the lowest score to language and grammar using a five-point scale. Furthermore, ChatGPT identified recurring writing issues, including a lack of clear thesis statements, underdeveloped arguments, insufficient specific examples or evidence, weak transitions and cohesion, unclear counterarguments, and grammar and syntax problems. These issues need to be reflected in writing instruction for improving argumentative essays. When it comes to the strengths of ChatGPT-generated feedback, it provided immediate and scalable feedback, explicit corrective feedback on language and mechanics, and a standardized overall evaluation that offered a comprehensive view at a glance. However, ChatGPT-generated feedback also had the limitations, including formulaic comments on content and organization, generic and overgeneralized feedback on final suggestions for improvement, and a lack of in-depth analysis of counterarguments and logical flow. The current study has pedagogical implications, offering valuable insights into the interpretation and analysis of ChatGPT-generated feedback as well as recommendations for its optimal integration within EFL contexts.
KW - ChatGPT-generated feedback;Artificial Intelligence;argumentative essays;Korean EFL classrooms
DO - 10.18095/meeso.2025.26.1.166
ER -
Oh, Hyun-Joo and Hsieh, Shu-Fei. (2025). An Analysis of ChatGPT-Generated Feedback on Argumentative Essays in Korean EFL Classrooms. Modern English Education, 26, 166-181.
Oh, Hyun-Joo and Hsieh, Shu-Fei. 2025, "An Analysis of ChatGPT-Generated Feedback on Argumentative Essays in Korean EFL Classrooms", Modern English Education, vol.26, pp.166-181. Available from: doi:10.18095/meeso.2025.26.1.166
Oh, Hyun-Joo, Hsieh, Shu-Fei "An Analysis of ChatGPT-Generated Feedback on Argumentative Essays in Korean EFL Classrooms" Modern English Education 26 pp.166-181 (2025) : 166.
Oh, Hyun-Joo, Hsieh, Shu-Fei. An Analysis of ChatGPT-Generated Feedback on Argumentative Essays in Korean EFL Classrooms. 2025; 26 166-181. Available from: doi:10.18095/meeso.2025.26.1.166
Oh, Hyun-Joo and Hsieh, Shu-Fei. "An Analysis of ChatGPT-Generated Feedback on Argumentative Essays in Korean EFL Classrooms" Modern English Education 26(2025) : 166-181.doi: 10.18095/meeso.2025.26.1.166
Oh, Hyun-Joo; Hsieh, Shu-Fei. An Analysis of ChatGPT-Generated Feedback on Argumentative Essays in Korean EFL Classrooms. Modern English Education, 26, 166-181. doi: 10.18095/meeso.2025.26.1.166
Oh, Hyun-Joo; Hsieh, Shu-Fei. An Analysis of ChatGPT-Generated Feedback on Argumentative Essays in Korean EFL Classrooms. Modern English Education. 2025; 26 166-181. doi: 10.18095/meeso.2025.26.1.166
Oh, Hyun-Joo, Hsieh, Shu-Fei. An Analysis of ChatGPT-Generated Feedback on Argumentative Essays in Korean EFL Classrooms. 2025; 26 166-181. Available from: doi:10.18095/meeso.2025.26.1.166
Oh, Hyun-Joo and Hsieh, Shu-Fei. "An Analysis of ChatGPT-Generated Feedback on Argumentative Essays in Korean EFL Classrooms" Modern English Education 26(2025) : 166-181.doi: 10.18095/meeso.2025.26.1.166
