| تعداد نشریات | 44 |
| تعداد شمارهها | 1,924 |
| تعداد مقالات | 15,640 |
| تعداد مشاهده مقاله | 47,747,476 |
| تعداد دریافت فایل اصل مقاله | 19,217,949 |
Technology-Mediated Corrective Feedback in Online EFL Instruction: A Multi-Framework Analysis of Teacher vs. AI-Generated Feedback | ||
| Applied Research on English Language | ||
| مقالات آماده انتشار، پذیرفته شده، انتشار آنلاین از تاریخ 01 آذر 1405 | ||
| نوع مقاله: Research Article | ||
| شناسه دیجیتال (DOI): 10.22108/are.2026.151015.2882 | ||
| نویسندگان | ||
| Farahman Farrokhi1؛ Mohammad Zohrabi* 2؛ Alireza Asltaleb Maghferat1 | ||
| 1Department of English Language & Literature, Faculty of Persian Literature and Foreign Languages, University of Tabriz, Tabriz, Iran | ||
| 2Department of English Language & Literature, Faculty of Persian Literature and Foreign Languages, University of Tabriz, Tabriz, Iran | ||
| چکیده | ||
| Despite the ongoing integration of online platforms and Artificial Intelligence (AI) into language education, corrective feedback (CF) in online EFL instruction remains underexplored. This study investigated CF types provided by EFL teachers and AI (ChatGPT) across elementary, intermediate, and advanced proficiency levels. Adopting a qualitative descriptive, multi-phase, and triangulated approach, the study examined 40.5 hours of interactions in nine online EFL classes selected through a combination of intensity and purposive sampling. The interactions were observed, video-recorded, and transcribed, and learner errors identified in these interactions were subsequently used as prompts for AI to elicit parallel feedback. CF episodes were categorized using a multi-framework coding scheme developed for the study, while semi-structured interviews with teachers and AI examined feedback decision-making and rationales. Data were analyzed using descriptive statistics, chi-square tests of independence, and qualitative thematic analysis. Findings showed that human teachers provided more CF instances and a greater diversity of CF types than AI, predominantly employing recasts, whereas AI predominantly employed explicit correction across proficiency levels. Interview findings highlighted the multifunctional nature of CF for enhancing competence and performance, proficiency- and error-sensitive tailoring, the complementary affordances of human- and AI-mediated CF, and the importance of complementing error correction with positive feedback to boost affective and attitudinal factors. Overall, the findings suggest that technology-mediated CF is best conceptualized as a complementary approach in which AI affordances and human pedagogical expertise work together to provide constructive feedback rather than positioning AI as a replacement for human teachers. | ||
| کلیدواژهها | ||
| online؛ platforms؛ language؛ education | ||
|
آمار تعداد مشاهده مقاله: 2 |
||