The effect of computer - based tests on students' writing performance

Writing is one of the skills in which most of EFL students have difficulty. To help students improve their writing skills, there has been a shift from traditional techniques to more innovative ones like integrating technology into writing instruction and these applications have yielded better results. However, in the assessment of writing, the advantages of technology have not been fully used. There are a few studies on whether students can perform better on computer-based writing tests. Therefore, the purpose of this study is to find out whether computer-based tests can improve students' performance in writing tests. The study adopted both qualitative and quantitative approaches. A quasi-experimental design was used as the research method. 36 prep class students at B1 level at Akdeniz University School of Foreign Languages participated in the study as control and experimental groups. In the beginning of the study, both of the groups took the same writing test on paper as pre-test. Then, every week for two months, the experimental group took an exam on computer while the control group took the same exam on paper. At the end of this period, the experimental group took the post-test on the computer and the control group did it on paper. The pre and post test scores of both groups were analysed through SPSS. According to results of the analysis, the effect of computer based test on students' writing performance was positive, though not statistically significant. Students in the experimental group were also asked to write a reflection on how they felt about taking computer-based writing tests. It was observed that they had positive attitudes towards this testing mode.