ADAPTIVE LEARNING PLATFORMS USING AI TO IMPROVE STUDENT ENGAGEMENT AND OUTCOMES

Abstract

The fast adoption of artificial intelligence in educational technologies has now given rise to the concept of adaptive learning platforms, which are supposed to personalize the learning experience, increase students engagement, and the performance thereof. This experiment will explore the efficacy of an adaptive learning platform with AI integrated against a non-adaptive digital learning platform, in terms of academic achievement, engagement, learning behavior, and retention. A quasi-experimental, mixed-methods design was employed to collect data related to 238 adaptive and comparison data students using the pre-and post-tests and engagement survey, learning analytics, retention test, and puppy feed-back. The results explain that the learners who utilized the adaptive platform scored much higher in the post-test and unit test, better transfer performance, and better learning retention in the long term. The results of engagement showed that the adaptive learners had better time-on-task, completion rate, voluntary practice, less frustration, and confidence. The analytics of learning also indicated a high productivity of the error-recovery behavior, such as improved usage of hints and reduced quitting behaviors following the making of incorrect answers. Sub group analyses showed that adaptive learning was beneficial to all students irrespective of the levels of low performers and especially high levels of performances. In general, the findings are indicative that AI-based adaptive learning systems could serve as viable instruction supports, but at the same time reinforce engagement and learning performance once conducted in compliance with the selected pedagogical and ethical factors.