Predicting Learning Achievement through Self-Regulated Learning Strategies, Motivation, and Programming Behaviors

With the growth of digital learning, there has been an increase in research related to learning analytics. Learning analytics can be used to identify potential problems and improve the quality of education by measuring, collecting, analyzing, and reporting data about the learners and their background to understand the learner's learning situation and learning environment. In addition, further analysis of students' learning behaviors can be used to provide adaptive and personalized teaching suggestions. This study aimed to analyze operational data from the coding process on an online learning platform, as well as data obtained from students' self-regulated learning strategy scales and self-regulated learning motivation scales. The objective was to investigate the correlation of these factors with students' academic achievement and to determine whether these factors can be utilized to predict student learning outcomes. The results of the study revealed a significant correlation between programming behavior and student grades. Variances in self-regulated learning strategies and motivation levels exhibit notable differences in academic performance. When incorporating these performance-related values as features in the development of predictors, it proves effective in forecasting students' learning outcomes. However, due to the limited sample size in this study, the predictive model may experience reduced accuracy or overfitting issues when applied to larger datasets. Therefore, for future predictions with larger samples, considerations should be made to adjusting model hyperparameters or modifying the features used to improve the accuracy of the predictions.