The Use of Google Translate for L2 Writing: Machine Translation and Errors of EFL Learners

Abstract

Learning English as a foreign language (EFL), students in Korea often resort to online machine translation (MT) tools to produce language output for speaking and writing. The purpose of the study was to identify the comparative advantages of using MT in EFL writing. To achieve this purpose, this study analyzed and compared errors in three types of English writings composed by Korean university students: direct writing in L2, self-translation, and translation via Google Translate (GT). The results of error analysis showed reduction in the quantity of most grammatical errors when machine translation was used to aid writing through translation. However, compositions generated with GT resulted in more lexical errors and sentence structure errors, implying the limitations of GT as well as possible increase in the breadth of expression for L2 writers. This study suggests EFL learners could use GT to experiment more in L2 writing with less concern about grammar errors. Also, instructors that make use of MT as a pedagogical tool in EFL writing classrooms should give special guidance for students on choosing words and expressions that are deemed appropriate in the English language. (185 words)