Abstract
Traditionally, algebra was a discipline reserved for Secondary Education, focused on symbolic manipulation and equation solving. However, its incorporation into early grades under the early algebra approach has consolidated into a robust field of research in Mathematics Education. Nevertheless, the assessment of Early Algebraic Thinking (EAT) in Primary Education often lacks psychometrically validated diagnostic instruments capable of capturing this holistic cognitive development. To address this gap, this study designed and validated an Early Algebraic Thinking Test (EATT) to evaluate the cognitive abilities of primary school students (ages 8–10). A set of 26 items was designed, derived from a literature review, operationalizing five phenomenological manifestations of algebraic thought into a continuous hierarchy based on cognitive demand. The instrument's content validity was refined by a panel of 15 experts who reviewed the clarity, consistency, and relevance of the items (${\rm S-CVI/Ave} = 0.96$; ${\rm S-CVI/UA} = 0.69$). The test was subsequently administered to 186 students aged 8-10. Construct validity was assessed via Confirmatory Factor Analysis (CFA) using the WLSMV estimator, confirming a robust unidimensional structure (${\rm CFI} = 0.996$, ${\rm TLI} = 0.996$, ${\rm RMSEA} = 0.013$). Item difficulty and person ability were jointly calibrated using the Rasch model. Demonstrating high categorical composite reliability ($\mu = 0.84$), the EATT provides a valid and reliable basis for identifying cognitive thresholds to guide pedagogical interventions aimed at developing higher-order mathematical skills from an early age.