<![CDATA[Assessment and evaluation of student learning outcomes are crucial aspects of education. Effective assessment requires well-designed evaluation test questions, one of them is multiple-choice questions. However, the quality of the multiple-choice questions made must be high. Quality questions are those that have undergone item analysis to assess the validity and reliability of the assessment results for students during the learning process. This study employed an evaluative research methodology with a descriptive quantitative approach. The research subjects involved 251 tenth grade students from two schools in Bombana Regency, Southeast Sulawesi Province. The evaluation was conducted using an instrument consisting of 10 physics questions that were created and validated with the assistance of teachers from each school. These questions covered the topics of the nature of physics, the scientific method, and work safety. The purpose of this study is to analyze the quality of the physics questions to determine whether they meet the criteria or not. Data collection was carried out using the QUEST software and analyzed using the Rasch model. The analysis results showed that the 10 questions had INFIT MNSQ values ranging from 0.77 to 1.33, which are consistent with the Rasch model. Furthermore, the OUTFIT t values for all item questions were obtained at values less than or equal to 2.00, indicating that the questions can be considered usable and passable. The reliability estimate score for the items is 0.98, indicating that this test has high consistency and can be relied upon to accurately measure students' learning outcome. Keywords: cognitive assessment, rasch modelling, Item response theory.DOI: http://dx.doi.org/10.23960/jpmipa/v25i2.pp671-685]]>
