The impact of machine translation on creativity in writing tasks in EFL classes

Due to the fact that there is a limited number of research on machine translation as a learning tool, it is still controversial whether it is effective or not. The aspect of those studies is also narrow. Most research approaches machine translation in terms of its impact on grammar and lexis or the perceptions of the users and the strategies they generated. However, a very important point is neglected: creativity. The concern of how to flourish creativity in EFL classes has long been investigated in the field of education. The current study combines the literature of these two different subjects – machine translation and creativity- and scrutinizes the impact of machine translation, Google Translate in this research, on creativity of beginner learners in writing tasks in EFL classes. It was conducted in a state Anatolian high school in Bursa, Turkey between September 2, 2019 and February 29, 2020 in the first semester of 2019-2020 educational year. It had a one-group quasi-experimental design with 35 beginner EFL (English as a foreign language) voluntary learners, whose parents had given a consent form. The study lasted for ten weeks including a training session, pre-and post-tests. Quantitative data were collected through a pre and post-test design and were analysed in terms of both creativity, grammatical and lexical errors with paired samples t-test and Wilcoxon test. The inter-rater coefficient alphas in the CAT analysis are found to be .86 for the pre-test and .85 for the post-test. In addition, the differences between the fluency, flexibility, originality and elaboration scores between the pre and post-test are statistically significant while no differences are detected between the grammatical and lexical error scores. Qualitative data were gathered through a semi-structured interview and the responses show that the students have a positive perception of implementing Google Translate in their writing classes.