An analysis of formal errors in a corpus of L2 English produced by Chinese students

This paper describes the investigation of a small corpus of writing of English for academic purposes produced by L1 speakers of Mandarin. The investigation involved the development of a tagset for the identification of formal errors in the corpus, and the subsequent analysis of these errors with a view to creating remedial grammar materials for Chinese students studying in the medium of English. Some past approaches to error analysis are discussed, the process of developing the tagging system is described, and error types are identified, categorised, quantified, described and (as far as possible) explained.