Several instruments that measure the teaching, learning, and assessment quality have been developed and published. However, a psychometrically sound instrument to measure teaching, learning, and assessment quality in early childhood care and education that suitable for the Malaysian context needs to be validated. Therefore, this study aimed to validate on teaching, learning, and assessment quality in early childhood care and education instrument, which contains 68 items. The sample comprised 3,498, selected by stratified random sampling from a population of all Malaysian kindergarten teachers. Data were analyzed based on the Polytomous Item Response Theory (IRT) using the Xcalibre software. Samejima's Graded Rating Model (SRGM) was found to be the fit model with the data. Unidimensionality assumption and local independence were tested using the exploratory factor analysis and were fulfilled. The instrument's reliability was overall very good ([alpha]=0.966) and the construct validity was also fairly fulfilled with the value of 58.17% total variance explained. Therefore, this instrument is suggested to be used as fairly to measure the quality of Malaysian early childhood care and education among teachers so that appropriate follow-up actions can be implemented towards the betterment of early childhood education quality.