An AI-based English Grammar Checker vs. Human Raters in Evaluating EFL Learners’ Writing

Abstract

Computer-assisted language learning has been popular among EFL learners and teachers. In particular, automated English grammar checkers have made a significant contribution to improving learner’s writing ability and reducing the workload of teachers as well. However, studies investigating the validation of the AI-based grammar checker are scarce. This study collected 40 writing samples collected from Korean EFL high school second graders. The grammatical errors from collected samples were analyzed by the AI-based grammar checker and human raters. The study also identified the error types that an AI-based English grammar checker produced by comparing the feedback messages to human-raters’ corresponding corrections. The grammar checker disclosed the wrong feedback, same and different feedback compared to the human raters. The results indicated that the program failed to detect the prevalent grammatical errors and sometimes provided incorrect feedbacks, while it showed several successful ones. Nevertheless, this study shed light on the possibility of application of AI-based computer programs in EFL classroom. In addition, this study will be of interest to the English language teachers who deal with a large amount of students’ writings.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 27 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.