Online individualized corrective feedback on EFL learners' grammatical error correction

• Published in: Computer assisted language learning Vol. 37; no. 7; pp. 1449 - 1477
• Main Author: Ko, Chao-Jung
• Format: Journal Article
• Language: English
• Published: Oxford Routledge 02.09.2024; Taylor & Francis Ltd
• Subjects: College students; Colleges & universities; Computer assisted language learning; English (Second Language); English as a second language; English as a second language instruction; English Language Learners; Error analysis; Error correction; Feedback; Foreign languages; Grammar; Grammatical agreement; grammatical error correction; Individualized corrective feedback; Problem solving; Second language writing; self-correction; Tense; text feedback; Verbs; writing; Writing systems; Written Language; Taiwan
• DOI: 10.1080/09588221.2022.2118783
• ISSN: 0958-8221

This study aimed to examine the impact of an online writing system (OWS) providing individualized corrective feedback (CF) on learners' self-correction of grammatical errors (GE). It consisted of two phases: the pilot and the formal phases. Four EFL (English as a Foreign Language) Taiwanese university students participated in the study at the pilot phase and 22 at the formal phase. An OWS was developed to provide individualized CF in text modes. The data was collected from learners' four written works, their interview transcriptions, and the teacher's observation at the pilot phase, and from learners' pre- and post-written works, three online works, and a questionnaire at the formal phase. The findings of the study showed that individualized CF provided by the system seemed more beneficial to self-correction of GE for those learners at a higher proficiency level. They applied feedback to self-correction and problem-solving, which enhanced their autonomy development. By contrast, the system seemed less beneficial to lower-level learners' self-correction of GE. In addition, individualized CF seemed to be most beneficial to subject-verb agreement error self-correction, and least beneficial to verb tense correction. Different feedback types and provision modes should be considered when using the developed system to deliver corrective feedback.