Direct and Indirect Data-Driven Learning: Affordances and Constraints

Research on language learners’ employment of corpus tools and techniques for language learning and use, namely data-driven learning (DDL), is a flourishing field (Gilguin & Granger, 2010). Over 200 empirical studies had been published to evaluate corpus use in language learning (Boulton & Cobb, 2017). Although there is no shortage of pedagogical advice and empirical evidence supporting the pedagogical value of corpus application, DDL has not become part of mainstream teaching and learning practices. One plausible explanation for the paradox concerns teachers’ apparent impression that incorporating DDL into classroom teaching necessarily involves learners working with corpus tools and techniques via computers. If they find such direct access distracting, time-consuming, or technically challenging, teachers would hesitate to adopt DDL in their classrooms. This inhibiting impression may have resulted from the paucity of studies on indirect DDL alternatives, especially research comparing the effectiveness of direct and indirect DDL. To address this gap, we report an experimental study designed to compare the effectiveness of a direct and an indirect approach to DDL in facilitating Chinese learners’ mastery of hedging in an undergraduate English-as-a-foreign-language writing class. The study adopted a pretest/posttest/delayed posttest randomized two treatment design. The two experimental groups received, respectively, direct and indirect DDL instruction in the use of hedges in English academic writing. The direct approach involved students in conducting guided searches and analyses of target hedges in online corpora, whereas the indirect approach featured the use of corpus-informed, paper-and-pen learning tasks. Within- and between-group analyses of performances on the three tests yielded evidence of the strengths and limitations of the two approaches. A survey revealed the participants’ favorable attitudes toward the incorporation of corpora in classroom teaching, and their perceptions of the affordances and constraints of DDL. The paper concludes with a discussion of the pedagogical implications of these findings.