Abstract This study presents the development and validation of a 24-item instrument designed to assess students’ digital skills, an essential competency in modern education. Grounded in a robust conceptual framework, the instrument captures key dimensions of digital literacy and was tested using cross-sectional data alongside advanced latent variable modeling techniques.The analytical methods applied included Confirmatory Factor Analysis (CFA), Exploratory Structural Equation Modeling (ESEM), bifactor-CFA, and bifactor-ESEM. Among these, ESEM yielded the best model fit, offering a nuanced representation of the multidimensional structure of students’ digital skills.The model demonstrated strong psychometric properties, including high internal consistency, solid construct validity, and measurement invariance across gender. Predictive validity was also confirmed through significant associations with relevant educational outcomes.These findings support the instrument’s application in large-scale assessments and policy initiatives aimed at improving digital literacy. The validated framework provides a foundation for evidence-based decisions in curriculum design, teacher training, and educational planning, and is well-suited for integration into broader SEM framework-based research.
Frashëri et al. (Fri,) studied this question.