Validation of the Meta AI Literacy Scale (MAILS) in Ecuadorian High School Students: Evidence of Validity, Measurement Invariance, and Latent Profiles

This study aimed to validate the Meta AI Literacy Scale (MAILS) in Ecuadorian high school students and examine latent profiles of Artificial Intelligence (AI) literacy. An instrumental, cross-sectional study was conducted with a sample of 971 students aged 14 to 19 years. Confirmatory factor analysis (CFA) supported a nine-factor correlated model with satisfactory fit indices (CFI = 0.953, TLI = 0.946, RMSEA = 0.053, SRMR = 0.058), high internal consistency (total α = 0.973; ω = 0.974), and scalar measurement invariance by sex and type of educational centre. Latent profile analysis (LPA) identified four differentiated profiles organised along a quantitative continuum of competence. Overall, the findings provide empirical evidence regarding the validity and utility of the MAILS in a sample of Ecuadorian high school students, providing a basis for future validations in other Spanish-speaking and Latin American contexts, and suggest that intentional academic use of the internet is associated with higher levels of AI literacy, although with small effect sizes. The validated instrument constitutes a resource for educational researchers and practitioners seeking to characterise AI literacy in secondary school populations and to inform evidence-based curricular interventions.