Clustering Analysis of Financial Literacy Among Students

Financial literacy is a critical skill that young adults need to operate in the world of more complex and digitally mediated financial situations. This paper uses unsupervised machine learning to classify financial literacy profiles of 401 undergraduate students using the OECD/INFE framework, consisting of financial knowledge, attitudes, behavior, digital literacy, and risky financial practices. K-Means clustering was carried out and assessed through Elbow Method, Silhouette Score and Davies-Bouldin Index after z-score standardization to guarantee compatibility with Euclidean distance. The k = 2 solution generated the best validation values, with the creation of two distinct student type: a High-Literacy Low-Risk group, with high financial aptitude and defensive utilization of online financial products, and a Low-Literacy High-Risk group with weaker literacy and a higher score on risky behavior (use of online loans and paylater services). The results have demonstrated that financial literacy among students is heterogeneous and strongly related to risk-taking behavior, which calls the need to implement specific financial education and behavioral reinforcement strategies.