A Novel Analysis of the Index of Learning Styles of Undergraduate Engineering Students (#1736)

Authors

Isaza Dominguez, Lauren Genith

Robles Gomez, Antonio

Pastor Vargas, Rafael

Abstract

This study analyzed the learning preferences of 73 undergraduate engineering students of a Mechanical Engineering class at the University of X. The students filled out Felderman-Silverman Index of Learning Styles (ILS) questionnaires during class time and their scores were tabulated into a dataset according to student number in relation to each of the four ILS learning dimensions (active/reflective, sensing/intuitive, visual/verbal, and sequential/global). This dataset was first analyzed by k-means clustering using the Elbow Method, the Silhouette Index, and the Calinski-Harabasz Index to determine the optimal number of clusters which was found to be 2. The optimal number of clusters was then applied to a hierarchical clustering analysis using Ward's Method. Each cluster was then analyzed statistically to determine the dominant learning styles in each cluster. All analyses were done using Anaconda 4.3 software. Students in Cluster 1 showed a preference for active and intuitive learning and students in Cluster 2 favored reflective and verbal learning, This analysis was a novel approach to determining dominant learning styles in a group of undergraduate engineering students. The findings highlight the potential for improved educational outcomes in engineering by aligning curriculum design with student learning preferences