Ai-Powered Tutoring Systems: Revolutionizing Individualized Support For Learners

Abstract

Background: Over the past few years, researchers have paid increasing attention to AI-based tools including tutors for learning. Nonetheless, there is still limited knowledge of their effectiveness, user satisfaction, and the challenges they present. This study intends to interrelate these two variables and measure the association between satisfaction, personalization of learning, and overall efficacy of the system.

Objective: The general objective of this research is to systematically assess the effectiveness of the AI-based tutoring systems, and the satisfaction levels of their users, and look for plausible enhancements through analysis of user and system relations.

Methods: A quantitative research design was employed and a structured survey questionnaire was able to collect data from 250 users of AI tutors. Some of the key constructs included in this study are satisfaction, personalized learning, user-friendliness, and the effectiveness of the system were measured using 5-point Likert scales. Data was analyzed using descriptive statistics, correlations, and regression. Reliability, normality, and multicollinearity were also examined. The former is because of Cronbach’s Alpha, the Shapiro-Will test for the latter, and the Variance Inflation Factor was used to examine multicollinearity.

Results: Participants responded even though there were imbalances in the sample population. The Shapiro-Wilk test results exhibited all p-values (for all variables concerned) under 0.05%. The findings of the Cronbach's Alpha coefficient in the questionnaire also yielded persistence at 0.175. From linear regression, it was possible to observe users’ satisfaction with guarantees, personalization of education, and effectiveness of the system as determinants with an r2 value of -0.009. The VIF scores are moderate in that they were in the range of 4.2 to 4.6.

Conclusion: There is a moderate level of user engagement and satisfaction with the AI-based tutoring systems but these factors do not strongly predict their effectiveness. There’s a significant area of concern regarding the internal consistency of the questionnaire, including bimodal distribution. In the next research, it’s recommended to include more parameters regarding the success of AI TUTORS: content, teachers’ motivation, etc.