Discourse differences between international teaching assistants and university registers: a quantitative corpus approach

Abstract

Adopting a multidimensional analysis framework, the purpose of this study is to increase our understanding of international teaching assistants’ (ITAs) instructional language use by investigating co-occurring patterns of their linguistic features and to evaluate the degree to which the discourse features elicited by ITA assessment correspond to those of the target language actually used in university settings. Under these purposes, the effects of ITAs’ English proficiency levels, gender, and disciplines on these discourse features are also examined. The data come from 186 prospective ITAs’ mock-teaching presentations in ITA assessment at a large research university. The presentations were transcribed, proofread, and built into an ITA corpus consisting of over 247,000 words, and over ninety linguistic features were annotated. Based on the four dimensions identified from T2K-SWAL, the discourse features of ITA mock-teaching were found to differ from those of university registers (e.g., classroom teaching) in manifold ways. ITAs’ discourse features also varied by proficiency levels, gender, and disciplines. The study provides implications of the findings and concludes with discussing limitations and directions for future studies.