Data-Driven Learning at the English Drafting Stage

Data-Driven Learning at the English Drafting Stage

Hsien-Chin Liou, Tzu-Wei Yang
DOI: 10.4018/978-1-7998-2591-3.ch014
OnDemand:
(Individual Chapters)
Available
$37.50
No Current Special Offers
TOTAL SAVINGS: $37.50

Abstract

Integrating corpus consultation to help students' re-use of vocabulary items for essay-drafting has not been sufficiently examined in instructional contexts. This chapter investigated the linguistic features of target words for which students consulted concordance programs to include in essays by documenting the DDL of three groups of EFL students in a Taiwanese college writing program. Participant essays, video files of consultation processes, responses to questionnaires, interviews, and written records of corpus consultation were examined. Researchers found 88.14% of all consulted words were incorporated into essays, with learners looking up 3.18 words. The lexical profile of consulted words indicates that learners mainly used the most common 1,000 to 2,000 words with more verbs more frequently consulted than nouns. Students queried words for confirmation or incorporated the discovered vocabulary into essays. As emergent pattern hunters, the participants made only limited use of dictionaries and bilingual resources. Implications of the findings are discussed.
Chapter Preview
Top

Introduction

Recent years have seen corpora and concordance programs as significant resources for general foreign language (FL) learning and teaching and as practical references for writing in particular. Corpora are formed with a large amount of textual data of language (recent developments include multilingual and multimodal corpora). Many publishers of dictionaries or textbooks advocate the use of corpora for editing and compiling of their products. Those approaches are usually grouped as indirect use of the language corpora. Data-driven learning (DDL, Johns, 1991), that is direct corpus use by learners to acquire linguistic features, is shown to enhance various aspects of language acquisition (Boulton & Cobb, 2017; Chen & Flowerdew, 2018; Crosthwaite, 2019; Pérez-Paredes, 2019). This is also known as learner concordancing (e.g., Liou, 2019). Under DDL, learners have direct access to computer programs with one or several corpora for consultation of lexical-grammatical features, which are called concordancers. Learners with such self-directed methods of observing a great amount of language input are expected to become researchers with inductive abilities of drawing patterns from careful analyses. On the other hand, dictionary users usually view no or a limited set of examples for headword entries but consult them mainly to obtain a sense of meaning. Research showed that a variety of corpus systems were exploited by learners for writing to enhance writing ability (Chen & Flowerdew, 2018). FL learners used concordancers not only for discovering unknown vocabulary while writing but also for error correction (Crosthwaite, 2017; Dolgova & Mueller, 2019; Liou, 2019) or other purposes.

Chen and Flowerdew (2018) examined 37 studies concerning using corpora to assist academic writing instruction. Few of them discussed the literature on using DDL for drafting in the pre-writing stage. The above authors assert that its development is still at an exploratory stage. To illustrate, corpus consultation has been observed and examined in assisting different aspects of writing: cultivating learners’ efficacy in drafting (Kennedy & Miceli, 2017; Wu, in press), revising errors (Crosthwaite, 2017; Dolgova & Mueller, 2019; Liou, 2019), or enhancing writing abilities (Cotos, 2014; Huang, 2014). Concordancers can serve L2 students in the pre-, during, or post-writing stages for various pedagogical purposes. Compared with its many documented applications, corpus consultation at the drafting stage of written texts is much less explored in various educational contexts, and research is even more limited for English-as-a-second/foreign-language learners (EFL). To bridge this research gap, the current action research study aims to investigate how FL learners can apply DDL to discover vocabulary for drafting their writing. The project analyzed the corpus consultation undertaken by three groups of undergraduate EFL students (freshmen and sophomores) for applying the vocabulary item they looked up into their drafts while writing. In this chapter, we hope to add to the literature on the use of DDL in English drafting.

Key Terms in this Chapter

Lexical profile: Lexical profile is a well-known and widely used lexical richness measure. Depending on which computer program the researcher has access to, BNC-COCA 25( Nation, 2012 ) can divide a text into 25 levels starting from the most common 1000 English words to 25,000 words.

Pattern hunting: (Kenney & Miceli, 2010 AU21: The in-text citation "Kenney & Miceli, 2010" is not in the reference list. Please correct the citation, add the reference to the list, or delete the citation. , 2017 AU22: The in-text citation "Kenney, Miceli 2017" is not in the reference list. Please correct the citation, add the reference to the list, or delete the citation. ): One specific type of application of concordance programs for writing is to explore one or various alternative words for the learner’s target vocabulary item in order to achieve the purpose of content enrichment.

Complete Chapter List

Search this Book:
Reset