2019年9月14日 第53回日本語教育方法研究会 (ポスター発表)

When teaching composition, it is important not just to point out mistakes, but also to understand each student’s pattern of errors as well as those common among the entire class. Co-Chu, the corpus analysis system being developed by the authors, lets researchers analyze data they have collected and allows metadata tagging not only at the word level, but also the sentence level. We investigated the use of this feature to analyze data collected from Japanese language learners’ compositions. In this experiment, we focused on the use of transitive and intransitive verbs, tagging them at the word level, and attached metadata as comments to accompanying words such as particles and auxiliary verbs. As a result, we were able to use Co-Chu’s search feature to create a summary of how both individuals and the entire class used transitive and intransitive verbs.