<![CDATA[Mid-semester summative assessments play a crucial role in supporting competency-based learning in the Kurikulum Merdeka. However, existing studies and field practices indicate a persistent gap: teachers rarely conduct systematic psychometric evaluations. Addressing this gap, this study aims to (1) analyze the structure and characteristics of a mid-semester mathematics summative assessment and (2) evaluate the quality of its items based on psychometric criteria within the framework of CTT. Using a mixed-methods sequential exploratory design, data were obtained from two mathematics education experts, two mathematics teachers, and a school principal in an Islamic Integrated Secondary School in Sukoharjo Regency. Data sources included interview transcripts, assessment documents, students’ response sheets, and expert validation forms. Qualitative data were analyzed through data reduction, display, and conclusion drawing, while quantitative data were examined using Aiken’s V and CTT. The findings reveal that the assessment consisted of 40 multiple-choice items and 5 essay questions, covering Number and Algebra elements of Phase D in the Merdeka Curriculum. The items' content validity was moderate, with strengths in language but weaknesses in cognitive level alignment. Empirical results showed some multiple-choice items were invalid, while all essay questions were valid and reliable (r = 0.88). Most items were moderately difficult, with a discrimination index from fair to excellent (0.3 ≤ D ≤ 0.8). However, nearly one-third of distractors in the multiple-choice items did not function well. These results highlight the need for improved item construction and teacher capacity-building to ensure assessments that align with the principles of the Kurikulum Merdeka and support high-quality measurement of student competency.]]>
Copyrights © 2025
