인공지능(AI) 기반 적응형 학습을 적용한 토익수업의 학업성취 효과 및 영향요인

Abstract

Objectives The purpose of this study is to derive implications for the effective design and operation of TOEIC classes applying artificial intelligence (AI)-based adaptive learning by examining the effect of these classes on academic achievement and influencing factors. Methods To verify the effect, the differences in the pre and post TOEIC scores of all students who participated in the AI-based TOEIC classes at S University and groups divided by their initial TOEIC level were analyzed through paired sample t-tests. To examine the factors influencing the effect, the differences in TOEIC score improvement were analyzed in relation to student background variables through independent sample t-tests and one-way ANOVAs. AI learning factors that influence TOEIC score were examined through simple linear regression analyses. A total of 76 students’ data were used for these analyses. Results As a result of the analysis, it was discovered that the post-TOEIC scores of all participating students and the groups which had been divided based on the pre-TOEIC score (upper/lower) were significantly higher than those of the pre-TOEIC. No student background variables (gender, grade, department, or initial TOEIC level) affected the TOEIC score improvement. Also, it was found that, out of the AI learning factors (attendance rate, learning time, number of problems solved), the attendance rate and the number of problems solved had a significant effect on the TOEIC score. Conclusions TOEIC classes with AI-based adaptive learning that consider individual students’ learning levels and speed can contribute to the realization of customized education for students. To enhance the effectiveness of TOEIC classes using AI, this study suggests the application of blended learning to support students’ self-directed learning and the design and operation of classes that promote practical learning, such as periodic learning and solving many problems.