This study aimed to validate an instrument for measuring digital skills among pre-service science teachers using the Rasch model framework. A total of 349 Generation Z participants were surveyed from three year levels. Data were collected through Google Forms and analyzed using Winsteps 3.7.3. The Rasch model provided insights into item reliability, person reliability, separation indices, unidimensionality, item fit statistics, and differential item functioning (DIF). Results demonstrated strong psychometric robustness, with person reliability of 0.95, item reliability of 0.98, and Cronbach’s alpha of 0.97. Wright Map analysis indicated appropriate alignment between item difficulty and respondents’ abilities. Item fit statistics fell within the acceptable range, supporting internal validity. Unidimensionality analysis confirmed that the instrument measured a single construct, while DIF analysis showed no significant bias across gender or year of study. Overall, the instrument proved reliable, valid, and equitable in assessing digital skills among science pre-service teachers, offering a valuable tool for teacher education and digital competence assessment.
Copyrights © 2026