From Human Raters to AI: A Critical Review of Research on ChatGPT in Writing Assessment
Futurity Education · 18.06.2026 · 6 (2) · 193-212
DOI: 10.57125/fed.2026.06.12 ↗Kurz gefasst
Fachartikel in der Zeitschrift Futurity Education, online erschienen am 18.06.2026. Verfasst von Yu-mei Wang und 2 weiteren Autor:innen. Die Arbeit nennt 70 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
ChatGPT holds great promise for assessing student writing, an area greatly in need of research and exploration. The purpose of the current paper is to synthesize studies that used ChatGPT to assess student writing, focusing on the following questions: (1) What is the internal consistency of ChatGPT in writing assessment? (2) How does ChatGPT assessment performance compare with that of human raters? and (3) What factors influence ChatGPT performance in assessing student writing? The findings of the paper provide important insights on accuracy, consistency, and efficacy of ChatGPT as a tool for writing assessment. This research is significant and timely as educational institutions continue to seek a pedagogically valuable and cost-effective method for assessing student writing.
Zitieren
APA
Wang, Y., Harmer, P. A., & Xue, C. (2026). From Human Raters to AI: A Critical Review of Research on ChatGPT in Writing Assessment. Futurity Education, 6(2), 193-212. https://doi.org/10.57125/fed.2026.06.12
Vancouver
Wang Y, Harmer PA, Xue C. From Human Raters to AI: A Critical Review of Research on ChatGPT in Writing Assessment. Futurity Education. 2026;6(2):193-212. doi:10.57125/fed.2026.06.12
BibTeX
@article{Wang2026,
author = {Wang, Yu-mei and Harmer, Peter A. and Xue, Changsong},
title = {From Human Raters to AI: A Critical Review of Research on ChatGPT in Writing Assessment},
journal = {Futurity Education},
year = {2026},
volume = {6},
number = {2},
pages = {193--212},
doi = {10.57125/fed.2026.06.12}
}
EUVIMED