Latent dimensions of financial behaviour derived by principal component and factor analysis

Abstract

Abstract Rather than directly measuring the dimensions of financial behaviour, this study constructs latent dimensions from observed indicators through multivariate analysis. These latent constructs are subsequently aligned with conceptually developed dimensions. Dimensionality reduction of the survey data informs the construction of the latent space. Data were collected using a tailored questionnaire administered to a sub-Saharan African sample (470 participants from urban and rural Ethiopia). Principal component analysis (PCA) and factor analysis (FA) were employed to test the framework and examine interrelationships among dimensions. The results provide exploratory support for aspects of the proposed multidimensional framework; however, further validation through confirmatory factor analysis (CFA), replication, or external behavioural outcomes is necessary. Financial inclusion emerges as a relatively distinct, access-related dimension associated with usage, but its status as a prerequisite for broader financial behaviour requires additional investigation. Financial literacy is most strongly linked to usage and knowledge-related indicators, and while its role in risk management remains conceptually plausible, it requires direct empirical testing. The findings highlight the limitations of inclusion-focused metrics and underscore the multidimensional nature of financial literacy. By integrating a novel conceptual framework, original data, and advanced statistical techniques, this study provides a comprehensive approach to understanding financial behaviour and challenges traditional assumptions, thereby establishing a foundation for future research and more effective strategies in financial education and inclusion.