<![CDATA[The development of the digital era and the increasing need for human resources who are adaptable to information technology have prompted universities to utilize academic data to improve the quality of learning. This study aims to develop a multiple linear regression model to predict students’ final grades in the Computer Architecture and Organization course based on learning evaluation variables. The predictor variables used include Quiz 1 score, Quiz 2 score, assignment score, Midterm Exam (UTS) score, and Final Exam (UAS) score, while the students’ final grades are set as the dependent variable. The study employs a quantitative approach involving data collection, data preprocessing, splitting the dataset into training and testing sets, constructing the linear regression model, and evaluating model performance using Root Mean Squared Error (RMSE), Mean Squared Error (MSE), and the coefficient of determination (R²). The dataset was split into 75% training data and 25% testing data. The results indicate that adding predictor variables significantly improves model performance. The best model was obtained by combining quiz, assignment, and midterm exam variables, with an RMSE of 2.30, an MSE of 5.28, and an R² of 0.81. These findings indicate that multiple linear regression is capable of predicting students’ final grades with a high degree of accuracy and can explain the relative contribution of each academic variable to students’ learning outcomes. This study is expected to support the implementation of learning analytics and data-driven decision-making in the evaluation of learning at the university level.]]>
