Coh-Metrix를 통한 한국인 EFL학습자와 영어 모국어 화자의 논설문 코퍼스에 나타난 언어적 특성 분석

The purpose of this study was to examine the linguistic features shown in the corpus of Korean EFL students and Native Speakers’ argumentative essays with a computer-based analyzing tool, Coh-Metrix. On the basis of YELC (Yonsei English Learner Corpus) and LOCNESS (Louvain Corpus of Native English Essays), 12 lexical and 10 syntactic variables were identified, with the consideration of English proficiency levels. The results indicated that lexical and syntactic variables appeared to cause differences among the three groups (HEFL, EFL students with high proficiency; LEFL, EFL students with low proficiency; NES, Native English speakers). Significant group differences were shown in lexical variables including the total number of sentences, the total number of words, mean sentence length, type-token ratio, word frequency, age of acquisition, familiarity, concreteness, imagability, polysemy, hypernymy for noun, hypernymy for verb. Also, Significant group differences were shown in syntactic variables including the number of words before main verb, the number of modifiers in noun phrase, and density of prepositions, negations, gerunds, and infinitives. This study suggested some implications for EFL writing theory and pedagogy.