한국어 말하기 자동 채점 모델 개발을 위한 설계 요소

Design Elements for Developing an Automated Scoring Model for Korean Speaking

This study aims to propose design elements to help develop automated scoring models in Korean speaking assessment. In language learning, automated scoring refers to the use of computers to assess and score language skills. Automated scoring has the advantage of reducing the burden on raters and enabling more reliable assessment and faster feedback. This study examined the current state of research on automated scoring for Korean speaking assessments, as well as the current situation in automated scoring for English speaking assessments. Based on this analysis, this study proposes several design elements for developing an automated scoring model for speaking assessments, including the purpose and type of evaluation, evaluation construct and scoring features, scoring scale, application type, task type, and system type. Furthermore, it argues that the development of such a model should comprehensively and specifically consider these factors. This study is expected to contribute to the development of an automated scoring model in Korean speaking assessment with validity, reliability, and feasibility.