How Does Data Driven Learning Affect the Production of Multi-Word Sequences in EAP Students’ Academic Writing?

Abstract

Abstract. Multi-word sequences (MWSs) have been found to occur with high frequency in academic writing [2,25]. MWSs are recurrent expressions, which a writer retrieves from his/her long-term memory in order to construct utterances. In written discourse, such MWSs serve to refer the reader to previous research, organize the sections of texts and discourse within them and position the writer as knowledgeable [2,3]. Previous research suggests that L2 writers frequently misuse these forms, resulting in disfluent written discourse [8,20,23]. Nevertheless, Hyland [14] suggests that use of appropriate and sophisticated MWSs helps to establish the writer as a member of an academic discourse community. The current, corpus-based and quasi-experimental study investigates the effectiveness of using Data Driven Learning (DDL) in conjunction with teaching MWSs. Key MWSs have been selected from Simpson-Vlach and Ellis' [25] Academic Formulas List (AFL), specifically from the Referential Function and two sub-categories of the Stance Function, Hedges and Epistemic Stance. The researcher used an objective pretest-posttest to ascertain how DDL affects students' receptive knowledge of AFL-MWSs and used the first and final drafts of an argumentative essay to assess students' ability to produce them. A statistically significant difference between pretest and posttest scores for the treatment group supports the assertion that DDL positively impacts students' receptive knowledge of AFL-MWSs. Discussion includes comparison between students' self-generated inductions regarding each AFL-MWSs and how they used them within their essays.