Abstract
There is broad agreement that computational thinking (CT) is a foundational problem-solving competence with cross-curricular benefits, yet the field lacks validated assessments to test these claims. Because many existing CT tests are based on programming tasks, it is unclear whether they measure genuine CT skills or programming experience. To address this, we developed the Radboud Computational Thinking Test (RCTT), an unplugged paper-and-pencil assessment that focuses on abstraction, decomposition, and algorithmic thinking. This preregistered validation study (N≈550 upper-primary pupils) uses confirmatory factor analysis, classical test theory, and item response theory (IRT), using a 1PL model, alongside convergent, discriminant, and fairness testing. The project aims to deliver a validated, publicly available CT assessment to support evidence-based CT education and research.

This publication has 13 references indexed in Scilit: