E-learning for Licensure Examination Preparation with Predictive Analytics

As more education technologies are being deployed in the field of education, the use of predictive analytics when it comes to e-learning environments is one of the key approaches to the improvement of the learners’ achievements. This study focuses on logistic regression and decision trees that exist and operate in an e-learning platform for Civil Engineering students preparing for licensure examinations. The practice review, practice examination, and performance forecast modules comprise the e-learning. A set of 1,000 students was employed to train and cross-validate the predictive models. A mean average precision of 94% was achieved with the logistic regression model, along with precision and recall rates of 91% and 97%, respectively. Visualization through the use of the decision tree model emphasized the impact of study habits, such as completing more than 10 exams of practice and more than 80% of modules, on passing probability. The e-learning platform was assessed using the FURPS model, with an overall weighted mean rating of 4.52, interpreted to mean "Excellent" per student participation. These findings highlight the focus on persistent practice, complete execution of study materials, and active participation in the process of learning towards attaining desirable test outcomes. Predictive analytics methods can thus enhance the performance of the student in an e-learning setup, such that performance using these technologies has been illustrated through the application of these technologies to an e-learning platform.