Predicting Factors Related to Student Performance Using Decision Tree Algorithm

Student performance serves as a vital gauge for assessing the quality of education delivered by educational institutions. However, it is subject to the influence of various factors. To unravel the intricacies of these influences, this study employs a decision tree algorithm via the RapidMiner application as a powerful data mining method. The decision tree approach not only adds precision to the analysis but also makes the interpretation of results more intuitive, thereby assisting educational stakeholders in making informed decisions. Our research aims to identify learning strategies that could enhance student performance by investigating the relationship between multiple factors, including gender, parental educational background, and test preparation on students' passing test scores. The findings reveal that participation in test preparation courses has a significant impact on students' likelihood of passing tests. In the absence of test preparation, other variables such as gender and parental educational background come into play. These insights can serve as a valuable reference for educational institutions, guiding them to mandate test preparation courses to increase students' odds of academic success. This study contributes to the existing literature by demonstrating the efficacy of decision trees in dissecting the multi-layered factors affecting student performance, thereby offering a robust analytical tool for educational research.