Comparing Explicit, Implicit, and Automated Written Corrective Feedback on Iranian Pre-Intermediate EFL Learners’ Writing Accuracy and Organization
Seyed Shoaib Mehdi Razaviir *
Department of Foreign Languages, Qo.C., Islamic Azad University, Qom, Iran.
Asghar Afshari
Department of Foreign Languages, Qo.C., Islamic Azad University, Qom, Iran.
Seyed Abdolmajid Tabatabaee Lotfi
Department of Foreign Languages, Qo.C., Islamic Azad University, Qom, Iran.
*Author to whom correspondence should be addressed.
Abstract
Aims: This study compared the effects of Explicit Written Corrective Feedback (EWCF), Implicit Written Corrective Feedback (IWCF), and Automated Written Corrective Feedback (AWCF) on the writing accuracy and organisation of Iranian pre-intermediate EFL learners.
Study Design: A quasi-experimental pretest–posttest design with three intact instructional groups was employed.
Place and Duration of Study: The study was conducted at Meraj Language Institute, Qom, Iran, over five weeks in 2025.
Methodology: Sixty CEFR A2 learners, selected through the Oxford Placement Test, participated in three groups of 20. All groups received the same materials, writing tasks, schedule, and classroom instruction; the planned difference among the groups was the feedback type. AWCF was generated using a standardised researcher-developed prompt with Google Gemini 1.5 Flash. Accuracy was operationalised as the mean of the Grammatical Range and Accuracy and Lexical Resource scores, whereas organisation was operationalised through Coherence and Cohesion, using an adapted IELTS-based 0–9 framework. Separate 3 × 2 mixed-design analyses of variance and follow-up gain comparisons were conducted.
Results: Accuracy showed a significant Time × Group interaction, F(2, 57) = 39.07, P < .001, partial η² = .578. Mean gains were 0.775 for EWCF, 0.525 for AWCF, and 0.225 for IWCF, with all pairwise comparisons significant. Organisation also showed a significant Time × Group interaction, F(2, 57) = 6.315, P = .003, partial η² = .181. Mean gains were 0.375, 0.175, and 0.075, respectively; only the EWCF–IWCF comparison was significant.
Conclusion: Feedback effects differed by writing dimension. EWCF showed the largest gains, particularly for accuracy, whereas AWCF produced intermediate accuracy gains and positive within-group change in organisation, with no significant difference from EWCF. The findings support a differentiated view of WCF that considers learner proficiency, writing dimension, and instructional context.
Keywords: Written corrective feedback, explicit feedback, implicit feedback, automated feedback, writing accuracy, writing organisation, pre-intermediate EFL learners