Case of Application of Essay-type Evaluation and Feedback Using Artificial Intelligence for Liberal Arts Education: Focusing on the basic liberal arts course ‘Understanding Sustainable Development Goals’

Abstract

The purpose of this study is to reveal the results of cases where artificial intelligence descriptive evaluation and feedback systems were applied in large-scale online liberal arts lectures at universities. For this purpose, we investigated and analyzed the results and satisfaction of descriptive evaluations conducted by students in the online liberal arts course ‘Understanding Sustainable Development’ taken by 2,600 students at University A. As a research method, frequency analysis and trajectory analysis were conducted to confirm changes in student participation rates, and factor analysis was used to cluster and compare students' narrative evaluation response trends. As a result of the analysis, first, more than 60% of the students who took the ‘Understanding Sustainable Development’ course participated even after the 14th week, and more than 70% of the students participated more than 10 times. Second, the students' evaluation scores showed a similar score range in the first week, but the range of scores increased as the 14th week progressed. In addition, there was a group whose score difference reversed around the 7th week, the midpoint of a semester's lectures, so it was possible to identify both the period and the group in which the students' learning needs to be carefully managed. Third, when looking at students' satisfaction with AI quizzes, 86.6% responded positively regarding their active participation in AI-based quizzes, and 83.1% responded positively regarding the usefulness of AI-based quizzes. Meanwhile, the degree of motivation and expansion of AI-based quizzes into other liberal arts classes was relatively low at positive rates of 52.6% and 61.5%, respectively, but was much higher than the negative rates of 19.6% and 11.8%. Through these results, this study confirmed that the artificial intelligence-based narrative evaluation feedback system that provides feedback to individual students has implications in terms of student learning management and quality management of liberal arts courses. In order to improve learners' learning fidelity and increase their academic retention rate in large-scale online liberal arts lectures in the future, interest in and improvement of the evaluation feedback system is essential, and its use can be expanded to other classes that require descriptive evaluation.