Students’ dependence on machine translation and their manual translation competence in translating descriptive texts

Abstract

This study is based on the increasing use of Machine Translation, particularly Google Translate and DeepL, in translation activities among EFL students. The ease and speed offered by these tools encourage students to use them frequently, which may affect their manual translation competence development. This study aimed to investigate students’ dependence on Machine Translation, their manual translation competence in translating a descriptive text, and the relationship between these two aspects. The participants of this study were 15 third-semester students from the English Education Study Program at Universitas HKBP Nommensen Pematangsiantar. Data were collected through a questionnaire on students’ dependence on Machine Translation and a manual translation test. In the translation test, students were asked to translate a descriptive text entitled “Universitas HKBP Nommensen Pematangsiantar.” Manual translations were assessed based on accuracy, vocabulary, and clarity. The data were descriptively analyzed using tables, percentages, and graphs. The findings show that students have a relatively high level of dependence on MT. Meanwhile, students’ manual translation competence was at a moderate level, with an average score of 59.1%, in which clarity was the strongest aspect and accuracy the weakest. In comparison, translations produced using Machine Translation achieved higher results, with an average score of 81.7%. However, the Machine Translation results were used only as supporting data for comparison. These findings indicate that Machine Translation can assist the translation process, but excessive dependence may hinder the development of students’ manual translation competence.