81006675

open access

Version of Record

Using Multivariate Statistical Techniques to Analyze the Writing of East Asian Learners of English

English (英語)

Learner Corpus Studies in Asia and the World

1

55-65

神戸大学国際コミュニケーションセンター

2013-03-23

2014-06-24

The primary purpose of Contrastive Interlanguage Analysis (CIA) (Granger, 1996) is to reveal first language factors that influence the development of a learner language by comparing the performance of learners who speak different native languages. In the present study, we use CIA to investigate differences in the linguistic features of writing by different L2 East Asian learners of English. Multivariate statistical techniques, namely correspondence analysis and hierarchical cluster analysis, were used to examine the use of a wide range of linguistic features, such as (a) vocabulary, (b) parts-of-speech, (c) semantic categories for the major word classes, and (d) grammatical characteristics, in the writing of East Asian learners of English from different countries, including Hong Kong, Taiwan, Korea, and Japan. Data on the frequency of each linguistic feature are crucial for analyzing differences between learner language groups. Samples were sourced from ICNALE (Ishikawa, 2011), a corpus of a million words writing samples from EFL learners, which is considered to be the largest East Asian composition database. The purpose of this study was to identify the linguistic features that can be used to discriminate between different East Asian EFL learners and to distinguish native and non-native speakers of English to build up a general picture of interlanguage use. We expect our findings to contribute to our understanding of learner language development from multiple linguistic perspectives.

紀要論文