Modeling digital competence in education: A bifactor-ESEM approach to general and domain-specific skills

Across European and Western Balkan countries, a persistent misalignment remains between graduates’ digital skill levels and the rapidly evolving demands of the contemporary workforce. Addressing this gap requires evidence-based interventions grounded in multidimensional and psychometrically rigorous assessments of digital competence. This study applies a bifactor exploratory structural equation modeling (B-ESEM) framework to evaluate university students’ functional digital skills using items from the Youth Digital Skills Indicator. The model integrates a general digital competence factor with four domain-specific dimensions, capturing both shared and unique variance across skill areas. Using data from 603 students, results indicate a strong general digital competence factor alongside meaningful domain-specific contributions. The model demonstrates full measurement invariance across gender and shows predictive validity for technology affinity and digital self-efficacy. These findings underscore the relevance of conceptualizing digital competence as both transversal and domain-sensitive when informing curriculum design and educational policy. The study provides a validated and flexible measurement framework that can guide future research and targeted digital skill enhancement initiatives.