Rasch Investigation of the CogAT Form 8 Verbal, Quantitative, and Nonverbal Domains in Grades 3 Through 5.

This study evaluates the psychometric properties of the Cognitive Abilities Test (CogAT) Form 8, a widely used assessment for identifying K-12 students eligible for gifted education programs, by applying dichotomous Rasch modeling to data from 1,689 third- through fifth-grade students in a large, diverse urban Texas school district during the 2022–2023 school year. Results indicate strong preliminary evidence supporting the validity, reliability, and construct unidimensionality of the verbal, quantitative, and nonverbal domains, with item difficulties appropriately targeted to the referred student sample, though some item misfit and potential ceiling effects were observed. The study underscores the utility of Rasch analysis for providing objective measurement and recommends cautious continued use of CogAT Form 8 in similar diverse populations, while highlighting the need for further research on generalizability, differential item functioning, and classification accuracy across demographic groups.