The impact of computational modeling on student success in algorithm and programming track courses

Programming courses, such as Object-Oriented Programming (PBO), Algorithms and Data Structures (ASD), and Design and Analysis of Algorithms (DAA) at Parahyangan Catholic University's Informatics Study Program (IF UNPAR), have shown relatively low passing rates and average grades. To enhance students' problem-solving abilities, IF UNPAR's 2018 Curriculum introduced the compulsory course, Modeling for Computation (PUK). This study aims to analyze the influence of PUK grades on students' academic performance in programming courses. We used linear regression modeling and Naïve Bayes classification to predict student grades. The results show that the regression model yielded a residual standard error between 15.43 and 28.15, while the Naïve Bayes model achieved a Root Mean Squared Error (RMSE) between 1.29 and 1.85. These findings indicate that PUK grades can serve as an early indicator of student success in programming courses, simultaneously supporting the integration of modeling and problem solving capabilities into the informatics curriculum.