17–19 May 2024
Meijo University Nagoya Dome Campus
Asia/Tokyo timezone

Teacher approaches towards using chatgpt to analyze a corpus of student writing samples

19 May 2024, 09:00
30m
DN 404 (North Building)

DN 404 (North Building)

Research Presentation (30 minutes) AI for Teaching DN 404

Speakers

Jeanette Dennisson (Institute of Science Tokyo) Michael Barr (Kyoto University of Foreign Studies)

Description

AI tools such as ChatGPT can be used to great advantage by teachers, classes, and individual students to improve the quality of Academic English writing skills while providing superior and more customized feedback within the parameters of specific writing assignments. The authors intend to showcase the various ways ChatGPT could support best teaching practices and successful methodologies for writing tasks based on the analysis of a corpus of student writing samples collected before the arrival of ChatGPT. Three academic years of student writing samples were analyzed. Undergraduate EFL students of different CEFR levels were instructed to write 100+ words within 10 minutes according to a specific writing theme or prompt. These writing samples were then collected, digitized, and anonymized before they were separately analyzed by each author using ChatGPT. The results of each analysis were compared and interpreted in order to understand trends in student writing which could direct future coursework. Such an objective analysis of student work using ChatGPT can be extremely beneficial in (1) identifying areas that need improvement, (2) evaluating individual student progress, and (3) directly informing ChatGPT-based methodology in the teaching process.

Keywords Writing tasks, Corpus, ChatGPT, Writing analysis

Primary author

Michael Barr (Kyoto University of Foreign Studies)

Co-author

Jeanette Dennisson (Institute of Science Tokyo)

Presentation materials

There are no materials yet.