IJIET 2026 Vol.16(3): 687-694
doi: 10.18178/ijiet.2026.16.3.2541
doi: 10.18178/ijiet.2026.16.3.2541
Comparing AI-Based and Peer-Based Feedback in Teaching the CaRS Model: A Quasi-Experimental Study on Postgraduate Academic Writing
Nurul Lailatul Khusniyah1, Lukman Hakim2, and Afif Ikhwanul Muslimin1,*
1. English Language Education Study Program, Faculty of Education and Teacher Training, Universitas Islam Negeri Mataram, Mataram, Indonesia
2. Islamic Education Study Program, Faculty of Education and Teacher Training, Universitas Islam Negeri Mataram, Mataram, Indonesia
Email: nurullaila@uinmataram.ac.id (N.L.K.); lukmanhakim@uinmataram.ac.id (L.H.); afifikhwanulm@uinmataram.ac.id (A.I.M.)
*Corresponding author
2. Islamic Education Study Program, Faculty of Education and Teacher Training, Universitas Islam Negeri Mataram, Mataram, Indonesia
Email: nurullaila@uinmataram.ac.id (N.L.K.); lukmanhakim@uinmataram.ac.id (L.H.); afifikhwanulm@uinmataram.ac.id (A.I.M.)
*Corresponding author
Manuscript received August 17, 2025; revised August 26, 2025; accepted September 29, 2025; published March 13, 2026
Abstract—This study investigates the effectiveness of Artificial Intelligence (AI)-generated and peer-based feedback in enhancing postgraduate students’ ability to apply the Create-a-Research-Space (CaRS) model when writing research article introductions. Although automated feedback has gained increasing attention, its impact compared with peer review in genre-based writing instruction remains underexplored. Employing a quasi-experimental design, the study involved 41 postgraduate students of Islamic education at an Indonesian state Islamic university. Class B (n = 20) received AI feedback via ChatGPT with structured CaRS-based prompts, while Class C (n = 21) engaged in peer review using a CaRS checklist. Data were collected through pre-test and post-test scores assessed with an analytic rubric, complemented by an open-ended perception survey. The results showed significant improvements in both groups (Artificial intelligence (AI) group: mean score 2.75 → 4.35; peer group: 2.33 → 4.19), with no statistically significant difference between them. Perception data revealed that students valued the clarity of AI feedback and the contextual relevance of peer comments, though both modes had limitations. The findings suggest that AI and peer feedback are comparably effective and can complement each other in supporting genre competence. The study highlights the importance of integrating AI tools into academic writing pedagogy while cultivating students’ feedback literacy to maximize the benefits of diverse feedback sources.
Keywords—Artificial intelligence (AI) feedback, peer feedback, Create-a-Research-Space (CaRS) model, postgraduate writing, academic writing pedagogy
Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).
Keywords—Artificial intelligence (AI) feedback, peer feedback, Create-a-Research-Space (CaRS) model, postgraduate writing, academic writing pedagogy
Cite: Nurul Lailatul Khusniyah, Lukman Hakim, and Afif Ikhwanul Muslimin, "Comparing AI-Based and Peer-Based Feedback in Teaching the CaRS Model: A Quasi-Experimental Study on Postgraduate Academic Writing," International Journal of Information and Education Technology, vol. 16, no. 3, pp. 687-694, 2026.
Copyright © 2026 by the authors. This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited (CC BY 4.0).