THE EFFECTIVENESS OF USING DEEPL AS A TRANSLATION TOOL ON THE ACCURACY OF ENGLISH TRANSLATIONS AMONG ENGLISH DEPARTMENT STUDENTS AT UNIVERSITAS BHINNEKA PGRI

Abstract

This study aims to examine the effectiveness of DeepL Translator as a tool for improving the accuracy of English translation among sixth-semester students of the English Department at Universitas Bhinneka PGRI. Utilizing a pre-experimental one-group pre-test–post-test design, this quantitative research involved 16 participants. The research instruments consisted of pre-test and post-test translation tasks designed to assess students’ translation accuracy before and after using DeepL Translator. These tasks were evaluated by two independent raters using a detailed scoring rubric.

Data analysis included descriptive statistics, normality testing using the Shapiro-Wilk test, and hypothesis testing using the Wilcoxon Signed-Rank Test, as the pre-test scores did not meet the assumption of normality. The results indicated a significant improvement, with the average score increasing from 59.31 (pre-test) to 83.44 (post-test). This difference was statistically significant, with a Wilcoxon significance value of .000 (p < 0.05). Furthermore, inter-rater reliability was high, as shown by a Cronbach’s Alpha of 0.877 and an Intraclass Correlation Coefficient (ICC) of 0.833, indicating strong consistency between raters.