A Corpus-Based Study of Dependency Relation Types in English Writing by Chinese Senior High School Students: Evidence from CLEC

Abstract

This study investigates the use of dependency relation types in English writing by Chinese senior high school students. Based on the ST2 sub-corpus of the Chinese Learner English Corpus, it aims to reveal the syntactic features of students’ English sentence construction and provide implications for English grammar teaching. Using Python-assisted dependency parsing and frequency analysis, this study examines the frequency and distribution of dependency relation types in the English writing of Chinese senior high school students and compares them with those in a native-English corpus, GUM, within the Universal Dependencies framework. The findings show that the most frequently used dependency relation types include nsubj, Root, det, case, advmod, obj, obl, and amod. Compared with the GUM, students use fewer case relations and show less syntactic flexibility, especially in complex structures. The study suggests that grammar teaching should focus more on prepositional structures, common syntactic patterns, and complex sentence construction.