Assessing Computational Thinking in Early Childhood: A Systematic Review of Instruments, Constructs and Psychometric Evidence

Abstract

ABSTRACT Background As computational thinking (CT) becomes an essential component of early childhood education, there is a growing demand for valid and developmentally appropriate assessment tools. Yet, while the availability of assessments has increased, researchers have yet to establish psychometric rigour and fairness in assessment design and implementation. Objectives This systematic review synthesises the landscape of early childhood CT assessments by examining the constructs measured, the implementation formats used, and the reported psychometric properties, including reliability, validity, and fairness. Methods Guided by PRISMA protocols, we analysed 30 studies (from 28 articles) that developed, validated, or adapted CT assessments for children aged 3–8. Results and Conclusions Most assessments focus on algorithms/sequences, often using paper‐based, selected‐response formats. While many studies reported evidence of content and construct validity, an improvement over prior reviews, evidence of criterion validity and fairness is scarce. Our findings underscore the need for developmentally appropriate and psychometrically rigorous assessments to ensure all children are assessed fairly and without bias.