<![CDATA[Background: Poor numeracy performance among Indonesian students highlights the need for assessment tools with strong psychometric properties that are also culturally relevant. This study examined the validity and reliability of a numeracy assessment instrument based on Javanese ethnomathematics for fifth-grade students at an Islamic elementary school (Madrasah Ibtidaiyah). Method: The article reports the quantitative validation and psychometric testing phases of a broader Research and Development (R&D) study based on the Borg and Gall model. Thirty test items were developed in multiple-choice, complex multiple-choice, true-false, and short-answer formats, covering content domains, cognitive processes, and Javanese cultural contexts. Content validity was evaluated using Aiken's V and the Intraclass Correlation Coefficient (ICC), while Rasch modeling was applied to examine the instrument's overall psychometric characteristics. Result: The findings showed that all stimuli met the content validity criterion (Aiken's V ≥ 0.80), and inter-rater reliability was excellent (ICC = 0.909). Rasch analysis confirmed unidimensionality (raw variance explained = 46.0%), all items demonstrated positive polarity, and only one item failed to meet the item-fit criterion. Item reliability in the limited trial was very high (0.96). These findings indicate that the instrument possesses adequate psychometric quality to measure primary school students' numeracy reliably within a Javanese cultural context. Implication: The validated instrument provides a culturally responsive tool for assessing primary school students' numeracy within a Javanese cultural context]]>
Copyrights © 2025
