A Corpus-Based Study on Korean High School Students’ Common Errors in English Writing

Abstract

The purpose of this study is to investigate the common errors in writing production made by Korean high school students. It aims to provide information on the areas that Korean high school students need to develop and to inform their educational needs. For this, high school students’ writing samples were collected and built into a learner corpus named Shinhyeon Corpus of High school Learners’ English Essays (SCOHLEE). The study identified frequent errors produced by the high school students adopting a computer-aided error analysis (CEA) technique using the Louvain Error Tagging Manual, and quantitative and qualitative analyses were conducted by using corpus tools (AntConc 4.0). The results revealed that a total of 3,368 errors were categorized into 7 major categories and 47 sub-categories, with article errors (GA) produced the most. Regarding English proficiency, mid-level learners produced a higher rate of errors and a wider range of error types. Additionally, the study showed a significant increase in the frequency of punctuation errors in the second semester. Overall, this study has pedagogical implications for lexis, grammar and writing instruction and the provision of error correction.