Issues in the development of the British Academic Written English (BAWE) corpus

Abstract

The British Academic Written English (BAWE) corpus is a collection of texts produced by undergraduate and Master's students in a wide range of disciplines, for assessment as part of taught degree programmes undertaken in the UK. The majority of the contributors to the corpus are mother tongue speakers of English, but, in order to be included in the corpus, each assignment had to be judged proficient by assessors in the contributor's discipline, regardless of the writer's mother tongue. The corpus contains, therefore, only texts that have met departmental requirements for the given level of study. University writing programmes are typically aimed at undergraduate and Master's students, and it would be useful for writing tutors to know more about student assignment genres and the linguistic features of successful writing at undergraduate and Master's level. However, most large-scale descriptive studies of academic writing focus on published or publicly accessible texts, or learner essays on general academic topics, probably because there are practical difficulties associated with collecting large amounts of well-documented student output. This paper charts the experience of collecting data for the BAWE corpus, highlighting the problems we encountered and the solutions we chose, with a view to facilitating the task of future developers of academic student writing corpora.