Validity and Reliability Analysis Using the Rasch Model in Developing Creativity Tests Instruments for Elementary School Students

This research was motivated by the need to develop a creativity test instrument that is effective in assessing and developing the creativity of students in elementary schools. The aim of the research is to develop a valid and reliable creativity test instrument for elementary school students based on Torrance's theory, which includes fluency, flexibility, originality, and elaboration. The method used is a survey with a cross-sectional study design. The number of participants in this research was 772 upper elementary school students from one of the districts in Indonesia. Data analysis used the Rasch Model via the Winsteps application Version 3.73. The unidimensionality results on the unexplained variance in 1st to 5th contrast value are less than 15%, indicating that the creativity test instrument measures a single creativity construct, not influenced by external factors. The results of the item measure analysis revealed varying levels of difficulty among the items, ranging from very difficult, difficult, easy and very easy items. All items fit with Rasch Model analysis, indicating the suitability of the instrument for assessing elementary school students' creativity. Summary statistical analysis indicated high reliability, with Cronbach's Alpha at 0.9, person reliability at 0.91, and item reliability at 1.00, confirming consistency in student responses and the quality of the instrument items. This shows the consistency of students' answers in the excellent category and the quality of the items in the instrument category is special. This valid and reliable creativity instrument functions as a valuable assessment tool for the independent curriculum program in developing learning programs to improve educational outcomes and creativity of students in elementary schools.