The influence of ai-assisted writing tools and digital feedback on EFL students’ linguistic accuracy and writing performance in higher education contexts

Abstract

Background: The integration of artificial intelligence (AI) into language education has accelerated markedly in recent years, reshaping how feedback is delivered in second-language (L2) writing classrooms.

Aim: This study examined the influence of AI-assisted writing tools and structured digital feedback on the grammatical accuracy of English as a Foreign Language (EFL) students in higher education.

Method: A quasi-experimental, pre-test/post-test control-group design was used, comparing an experimental group receiving AI-based feedback (n = 36) with a control group receiving conventional teacher feedback (n = 36) across a six-week intervention. Grammatical accuracy scores were obtained at pre-test and post-test for both groups.

Results: The experimental group improved from a pre-test mean of 70.5 to a post-test mean of 84.2 (SD = 5.4), while the control group improved from 71.1 to 76.1 (SD = 6.0). An independent-samples t-test indicated a statistically significant between-group difference favoring the experimental group (t = 3.87, p < .05). These findings suggest that AI-assisted feedback was associated with a larger gain in grammatical accuracy than conventional teacher feedback over the same intervention period.

Conclusion: Fluency and revision-process outcomes originally planned as secondary measures are not yet available and are flagged as such rather than estimated. The results are discussed in relation to prior automated-feedback research and the broader debate on the relative contributions of AI-generated and human feedback in second-language writing instruction, together with the methodological caveats that necessarily accompany a two-group comparison of this scope.