<![CDATA[Assessment of science learning outcomes in junior high school science education is still largely dominated by conventional tests that do not adequately accommodate differences in students’ abilities and provide limited diagnostic feedback. This limitation highlights the need for a more adaptive and data-driven assessment system. This study aims to map the needs for developing a web-based Computerized Adaptive Test (CAT) system integrating Artificial Intelligence and Item Response Theory (AI-IRT) for science assessment at the junior high school level. The study employed a descriptive qualitative approach at the Analysis stage of the ADDIE development model. Data were collected through semi-structured interviews with one junior high school science teacher and five students, and analyzed using NVivo through open coding, axial coding, and selective coding. The findings reveal six key needs: adaptive assessment systems, measurement fairness and precision, diagnostic feedback, efficiency and automation, infrastructure readiness, and user-friendly interface design. These findings demonstrate the potential of integrating AI and IRT to support more accurate, personalized, and efficient science assessment. The results provide a conceptual and operational foundation for developing an AI-IRT-based CAT web system tailored to junior high school science learning.]]>
Copyrights © 2026
