Learning styles among older adults: an exploratory study through the Chaea instrument in the Pumuo program

Abstract

University dropout is a multifactorial phenomenon associated with academic, personal, and learning-related variables. This study analyzes the relationship between learning styles, metacognition, and university dropout within an e-learning course at an Italian university. An exploratory study was conducted with 257 students, utilizing 22 predictor variables derived from the CAMEA40 questionnaire, educational trajectories, personal characteristics, and academic records. Six classification models were trained: logistic regression, support vector machines (SVM), K-nearest neighbors, Random Forest, Naive Bayes, and multilayer perceptron. Model performance was evaluated using accuracy, precision, sensitivity (recall), and the F1-score. Random Forest achieved the best overall performance, with an accuracy of 0.91, precision of 0.91, sensitivity of 0.83, and an F1-score of 0.86. Although these findings require validation with broader samples, they suggest that predictive analytics can effectively support early detection and academic mentoring.