DocuScope, multi-dimensional analysis, and student writing

Abstract

In this chapter, we present a method for comparing tagging systems and patterns of disciplinary variation in corpora of student writing. We begin by highlighting the affordances of rhetorically and linguistically informed tagging systems by highlighting similarities and differences in each system’s analysis of the Michigan Corpus of Upper-Level Student Papers (MICUSP). Results confirm that both taggers produce statistically robust results in distinguishing disciplines across three dimensions and also highlight commonalities and differences that reflect the taggers’ respective theoretical orientations. Then, we present the results of a DocuScope-driven comparison of the British Academic Written English (BAWE) and MICUSP corpora and summarize topical and rhetorical patterns of disciplinary writing that seem fairly stable across national contexts. This chapter’s findings should prove useful to scholars interested in comparative methodologies of corpus analysis and rhetorical measures of disciplinary variation as well as those who work in or research writing in the disciplines.