A study on Vietnamese–English translation errors made by Google Translate as classified by text types
Studies in English Language and Education · 30.09.2026 · 13 (3) · 1550-1565
DOI: 10.24815/siele.v13i3.1663 ↗Kurz gefasst
Fachartikel in der Zeitschrift Studies in English Language and Education, online erschienen am 30.09.2026. Verfasst von Phuong Thi Lan Nguyen und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
While Google Translate (GT) is widely used in Vietnamese–English communication, its translation errors across different text types remain insufficiently understood. This study investigates errors made by GT in Vietnamese–English translation classified by text types. Specifically, it examines error frequency, common error types, and the most error-prone text type. A linguistic corpus consisting of 26 Vietnamese texts (totaling 5,740 words) classified into three types (expressive, informative, and vocative) is used as the input for GT’s translation into English. Three kinds of analyses were used in the analytical procedure, including translation analysis to analyze the source texts, translation comparative analysis to compare the source and target texts to identify errors, and statistical techniques to quantify the error frequency. The findings reveal that GT has made a smaller number of errors than indicated in previous studies on the English–Vietnamese language pair; however, it still makes 103 errors (an average of one error per 55.7 words), of which semantic errors are the most common (36 errors), far higher than lexical (26) and orthographic (20), while morphological, syntactic, and pragmatic errors are the least common with 7 errors each. Of the three types, vocative texts have the highest number of errors with 44 errors, followed by expressive texts with 40, while informative texts have the lowest error frequency (only 19). Based on research findings and discussions, implications have been made on how GT should be used, how its translation quality can be improved, and how translators should respond to machine translation.
Zitieren
APA
Nguyen, P. T. L., & Vu, H. T. (2026). A study on Vietnamese–English translation errors made by Google Translate as classified by text types. Studies in English Language and Education, 13(3), 1550-1565. https://doi.org/10.24815/siele.v13i3.1663
Vancouver
Nguyen PTL, Vu HT. A study on Vietnamese–English translation errors made by Google Translate as classified by text types. Studies in English Language and Education. 2026;13(3):1550-1565. doi:10.24815/siele.v13i3.1663
BibTeX
@article{Nguyen2026,
author = {Nguyen, Phuong Thi Lan and Vu, Hanh Thi},
title = {A study on Vietnamese–English translation errors made by Google Translate as classified by text types},
journal = {Studies in English Language and Education},
year = {2026},
volume = {13},
number = {3},
pages = {1550--1565},
doi = {10.24815/siele.v13i3.1663}
}
EUVIMED