Investigating the Chinese and English Language Proficiency of Tertiary Students in Hong Kong: Insights from a Student Translation Corpus

This study aims to investigate the Chinese and English language proficiency of tertiary students in Hong Kong through the unique lens of translation. An error-annotated translation learner corpus—the Hong Kong Student Translation (HKST) corpus, a subset of the Multilingual Student Translation (MUST) initiative—was developed following the standard of an international multilingual corpus initiative for the study of translated language of language learners and translation students. Consisting of Chinese‒English and English‒Chinese translations of students from more than 11 tertiary institutions in Hong Kong, the corpus includes over 300,000 word tokens. Analysis of the corpora data suggests that distortion in content transfer was the highest-frequency error type amongst students in both Chinese‒English and English–Chinese translations. Under the category of language errors, the top three problems in students’ English–Chinese translation were heavy structure (style and situational context), multiword non-term collocation (lexis and terminology), and pronoun reference (cohesion), and the top three in Chinese–English were tense/aspect (grammar), spelling (mechanics), and punctuation (mechanics). A close examination of these problems in relation to students’ individual differences were further explored. These findings pinpoint the most urgent problems that need to be addressed in both secondary and tertiary level of language teaching in Hong Kong. The study also indicates that tailor-made exercises should be developed to help enhance these identified deficiencies in students’ written Chinese and English language respectively.