Identifying Grammatical Errors and Mistakes via a Written Learner Corpus in a Foreign Language Context

Foreign language learners of English have difficulty in applying grammar rules in writing despite prolonged training focusing on grammar. This corpus-driven error analysis study examines English as a foreign language (EFL) learners’ grammatical errors through a written learner corpus, which contains essays written by Level 2 and 3 students in a language program at a state university. The study also aims to reveal whether they make any improvement within a term. Using James’s (1998) taxonomy of errors, the data were analyzed via a corpus tool, “AntConc”. The results of descriptive analysis for error frequency showed that the most common grammatical errors were of verb conjugation, prepositions, articles, grammatical numbers, and voice, respectively. The study also showed no significant progress for Level 2 learners while Level 3 learners slightly improved by rectifying the number of errors.