Which Feedback Is More Effective? A Comparative Study on Teachers' Evaluation of Feedback Models in Essay Writing

How Well Does AI-Generated Feedback Work? Intrinsic and Extrinsic Evaluation across more than 20,000 EFL Essay Drafts

This study examines feedback in English as a Foreign Language (EFL) writing contexts, focusing on written corrective feedback (WCF). Large language models (LLMs) can provide WCF at scale, but aligning them with pedagogical best practices remains an ongoing challenge. WCF meeting criteria like factuality or relevance may still be unsuitable for learning contexts, highlighting the need for extrinsic evaluation based on the learner's perspective. We deployed WCF systems in a university-level EFL class with nearly 2,000 students, collecting over 20,000 drafts. We evaluated the generated WCF from two perspectives: intrinsic evaluation by experienced English teachers using a rubric, and extrinsic evaluation via student feedback and engagement metrics. Results revealed low alignment between teacher expert ratings and student feedback. These findings suggest that traditional expert evaluation alone may not fully capture WCF's usability or helpfulness from the learner's perspective, highlighting the importance of learner-centered evaluation frameworks for AI-based applications in language education.

Teacher feedback and peer feedback: Benefits and challenges in EFL students’ academic writing development

Feedback plays a crucial role in language teaching and learning in general and development of academic writing skill in particular. Both teacher feedback and peer feedback play crucial roles in fostering writing skills, promoting learner engagement, and fostering self-regulation of learning. However, each type of feedback offers its benefits and challenges for EFL students’ academic writing development. Moreover, there are few published research studies in Vietnamese tertiary context related to comparing between the two types of feedback.  Using narrative review methodology, this study collected all evidence from a number of databases (ProQuest, ScienceDirect) to identify related studies for the current review. This narrative study aimed to provide deeper insights into the benefits and challenges of each type of feedback. The findings suggested recommendations to optimize feedback in order to effectively develop academic writing skills in EFL students in Vietnam. Furthermore, the study also offered the potential strategies for the best way of giving feedback in the AI era.

Feedback Literacy within Teacher Feedback vs. AI Feedback

This study aims to investigate how the age of artificial intelligence reshapes the world as numerous individuals get feedback. Regarding this issue, exploring how students perceive, interpret, and utilize feedback given by their language instructors and AI-driven tools gains vital importance. Methodology employs a descriptive design, wherein data are collected from 52 undergraduate EFL students at a school of foreign languages at a state university in Adana-Türkiye. The data includes writing parts of the Module 2 Exit Exam, reflective diaries of the students who get feedback from their language instructors, and those who get AI-generated feedback. The survey responses are also submitted in the study in 2025-2026 Academic Year. Two types of reflective diaries are collected via e-mail organized by the researcher. Quantitative data are analyzed to identify frequencies of feedback types and their reliance with exit exam, while qualitative insights are derived from student reflections via feedback diaries. The findings reveal that students reveal growing awareness of feedback processes. They tend to value immediacy and clarity in AI feedback while still relying on human instructors for contextual and affective dimensions. The study assumes to contribute to the field by situating feedback literacy within the rapidly changing digital landscape in pedagogical implications for integrating AI-supported tools into EFL writing instruction.

Students’ mathematical feedback processes using a digital curriculum resource: a qualitative perspective

This qualitative case study investigates how automated feedback from a widely accessible digital curriculum resource influences students’ mathematics learning. Drawing on a dialogic feedback framework, the study analyzes students’ feedback processes in an ecologically valid learning context. Grounded in a sociocultural perspective, the framework integrates Carless’ dialogic conceptualization of feedback, Shute’s feedback typology, and Vergnaud’s scheme theory. Eight eighth-graders from a German secondary school worked individually on percentage problems at home, video-recorded via online conferencing. Data included screen recordings and think-aloud protocols, which were analyzed using qualitative text analysis. Results reveal large individual differences in feedback processes—students vary in when and how they access feedback messages and how they make sense of them. Successfully solving a task after feedback does not always indicate understanding; in some cases, feedback aided procedural completion without conceptual insight. The effectiveness depended on the alignment between task design, feedback message content, and students’ actual difficulties. The study concludes that qualitative analysis of feedback processes with digital technologies in ecologically valid settings is necessary to target specific student difficulties and promote learning.

When AI-Mediated Feedback Writing Works and When It Does Not: Lessons Learned from Two Classroom Cases

The COM Essay Assessor

The development and calibration of the COM Essay Assessor is presented, a rubric-based generative artificial intelligence (GenAI) tool designed to support formative feedback while retaining instructor oversight and reflects on the opportunities and challenges of integrating GenAI into large writing programs.