ChatGPT-supported formative assessment in interpreter education
Cadernos de Linguagem e Sociedade · 29.03.2026 · 26 (2) · 265-290
DOI: 10.26512/les.v26i2.59793 ↗Kurz gefasst
Fachartikel in der Zeitschrift Cadernos de Linguagem e Sociedade, online erschienen am 29.03.2026. Verfasst von Wenjing Liu und 1 weiteren Autor:innen. Die Arbeit nennt 69 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Formative assessment and feedback play a critical role in teaching and learning. Recent advances in large language models (LLMs) have enabled their application as automated assessment tools and feedback providers. This study explores the validity of ChatGPT-based assessment and the perceived quality of its feedback in interpreter education. To this end, ChatGPT-4o was used to assess 60 Chinese–Portuguese simultaneous interpreting tasks, producing rubric-based quantitative ratings and qualitative feedback. To examine its effectiveness, three types of validity (concurrent, predictive, and know-group) were examined by comparing ChatGPT-generated scores with those of nine trained human raters. A post-hoc questionnaire was also administered to collect raters’ subjective perceptions of the feedback. Results show strong alignment between the model and human scores, with ChatGPT demonstrating robust predictive power and discriminative ability. Raters viewed the feedback favorably and supported its use as a complement to teacher feedback, highlighting the pedagogical value of LLMs in interpreter training.
Zitieren
APA
Liu, W., & Pagano, A. S. (2026). ChatGPT-supported formative assessment in interpreter education. Cadernos de Linguagem e Sociedade, 26(2), 265-290. https://doi.org/10.26512/les.v26i2.59793
Vancouver
Liu W, Pagano AS. ChatGPT-supported formative assessment in interpreter education. Cadernos de Linguagem e Sociedade. 2026;26(2):265-290. doi:10.26512/les.v26i2.59793
BibTeX
@article{Liu2026,
author = {Liu, Wenjing and Pagano, Adriana Silvina},
title = {ChatGPT-supported formative assessment in interpreter education},
journal = {Cadernos de Linguagem e Sociedade},
year = {2026},
volume = {26},
number = {2},
pages = {265--290},
doi = {10.26512/les.v26i2.59793}
}
EUVIMED