The Effect of a Diagnostic System for Scientific Competency Combined with a Personalized Intelligent Tutoring Module through Machine Learning for Enhancing the Learning Progression of Seventh Grade Students

Abstract

This study investigated the effects of a diagnostic system for scientific competency, combined with a personalized intelligent tutoring module that utilized machine learning, on the learning progression of seventh-grade students. The study employed a decision tree algorithm in Python to analyze secondary data from 847 students. The decision tree algorithm was used to create a predictive model that could diagnose individual scientific competency levels, which informed the development of individualized learning pathways. The experimental study involved 176 students, who were divided into two groups: an experimental group that received interactive feedback and a control group that did not receive any feedback. The study utilized a one-way MANOVA to evaluate three dimensions of scientific competency: Explanation of Scientific Phenomena, Evaluation and Design of Scientific Inquiry, and Interpretation of Data and Use of Evidence. The findings demonstrated significant improvements in the experimental group across all dimensions, highlighting the importance of immediate feedback in enhancing comprehension and motivation. The system was also highly rated for its quality and user satisfaction. This study emphasizes the potential of intelligent tutoring systems to improve scientific competency and suggests further educational applications.