DDL and learner corpora - CorpusCALL SIG Symposium

Research output: Contribution to conferencePaperpeer-review


Ever since the pioneering work on the development of the International Corpus of Learner English (ICLE) (Granger, 1993) in the late 1980s learner corpora have been compiled and served many purposes in language teaching and learning. The list of “Learner corpora around the world” maintained by the Centre for English Corpus Linguistics includes learner corpora of many languages with primarily written corpora (64%). Learner corpora have been used in testing and in materials design, and often focused on typical errors, overuse and underuse of specific linguistic items. Although a number of learner corpora have been compiled over the last decades, there are still only a few applications in data-driven learning (DDL). In line with this year’s conference theme of learner data, the symposium will focus on this lesser applied aspect of learner corpora. In addition, the symposium aims to be a platform to initiate discussion around the effective application of learner corpora in DDL.
The first paper by Luciana Forti (University for Foreigners of Perugia, Italy) and Reka R. Jablonkai (University of Bath, UK) will provide a review of previous studies that used learner corpora for DDL activities. As shown in Boulton & Vyatkina’s 30-year survey on DDL (Boulton & Vyatkina, 2021), the use of learner corpus data in DDL represents a minority, covering a mere 4.30% of the total of 489 studies reviewed by the authors. The presentation will seek for the reasons for this scarcity and will give an overview of the aspects in DDL language teaching that learner corpora have focused on.
The second paper by Tanjun Liu (Xi'an Jiaotong-Liverpool University) is to report an ongoing project focusing on the use of the data-driven learning (DDL) approach to enhancing students’ business writing in English. Twenty Chinese Business-majored undergraduates participated in the DDL hands-on activities by using a learner corpus of business writing and a corpus of expert business writing. Data from learners’ writing pieces and questionnaires on their perceptions were analysed. The preliminary results show that learners demonstrated an increase in the frequency, accuracy and variety in their use of collocations in writing after the DDL activities, and that they also benefited from the use of learner and expert writing data.
The third and last paper by Sandra Götz (Philipps University Marburg) focuses on the synergy between learner corpora and DDL. Despite the fact that “learner corpus research opens up exciting pedagogical perspectives (…) including classroom methodology” (Granger 2003: 542), so far, the overwhelming majority of studies on DDL have been using native speaker corpora, while learner corpora are still at the periphery of being used in DDL activities (albeit, see, e.g. Cotos 2014; Gablasova et al. 2019). In this conceptual paper, the benefits and drawbacks of using learner corpora in DDL in comparison to native speaker corpora will be systematically discussed.
Original languageEnglish
Publication statusAcceptance date - Aug 2022
EventEuroCALL 2022: Intelligent CALL, granular systems and learner data - Vigdis Finnbogadottir Institute of Foreign Languages , Reykjavik, Iceland
Duration: 16 Aug 202219 Aug 2022


ConferenceEuroCALL 2022
Internet address


  • data-driven learning
  • learner corpus

ASJC Scopus subject areas

  • Language and Linguistics


Dive into the research topics of 'DDL and learner corpora - CorpusCALL SIG Symposium'. Together they form a unique fingerprint.

Cite this