The impact of automated writing evaluation on English as a foreign language learners' writing self-efficacy, self-regulation, anxiety, and performance

Background

With the growing trend of integrating technology into teaching environments, using Automated Writing Evaluation (AWE) in writing instruction has been extensively studied over the last two decades. The studies on AWE mostly investigated its impact on students' writing proficiencies and revealed conflicting results. However, very few studies have investigated how the use of AWE impacts learners' writing-specific psychological factors, such as self-efficacy, self-regulation, and anxiety, from an empirical perspective. Therefore, further experimental studies are needed in this regard.

Objectives

This study aimed at examining how a combined automated-teacher feedback condition in which the students received automated feedback on their sentence-level errors alongside teacher feedback on content and organization impacted English as a Foreign Language (EFL) students' self-efficacy, self-regulation, anxiety, and writing performance, compared with a traditional teacher-only feedback condition.

Methods

The study utilized a quasi-experimental design in which two intact classes were randomly assigned to the experimental and control groups. During one semester (16 weeks), students in the experimental group received both automated and teacher feedback, referred to in this study as combined automated-teacher feedback. In contrast, students in the control group received only teacher feedback. Two scales, a questionnaire and a focus group interview, were used to obtain the data. The data were analysed using both quantitative and qualitative techniques.

Results and Conclusion

The quantitative results demonstrated no statistically significant difference between the two types of feedback procedures in promoting the use of self-regulated writing strategies by the students. However, the use of combined automated-teacher feedback was found to be more effective than conventional teacher-only feedback in enhancing the students' writing self-efficacy and writing performance. The two different feedback processes were shown to be ineffective at helping the students' writing anxiety, however. The qualitative data also showed that students had favourable opinions of their experiences with receiving automated and teacher comments together.

Implications

Considering the findings of this study regarding the impact of combined automated-teacher feedback on students' writing self-efficacy and writing performance, as well as positive student perceptions, it is suggested that this feedback model is used as an alternative to teacher-only feedback in EFL writing instruction.