Satisfaction with Research-Based Learning and Academic Performance by Big Data Analysis

Data science analysis was used to investigate the association between Research-Based Learning (RBL) and student’s academic performance. The research professors certified in RBL participated in this research. The participants used an advanced platform with over 500 resources. Tutors guided students in designing, implementing, and evaluating the RBL. We employed a quantitative and quasi-experimental design without a control group, correlational analysis, and ex post facto elements. A satisfaction survey was administered after the capacitation. In total, 342 teachers were trained and received accreditation. The analyses incorporated non-conventional data science methods such as correlation analysis, principal component analysis, and unsupervised learning utilizing the Density-Based Spatial Clustering Application with Noise algorithm. The professors’ satisfaction with RBL was related to student’s academic performance, which was found in clustering analysis.