To improve the quality of learning, learning evaluation has an important role. One of the learning evaluations that is often carried out is daily assessments to see the achievement of particular material. The aim of this research is to describe the quality of mathematics questions in statistics material for junior high school students. The method used in this research is quantitative descriptive research assisted by the AnBuSo program and R studio software. The data collection technique used documentation techniques, with data sources in the form of mathematics questions on statistics material. Question items analysis is applied to prove validity, estimate reliability, level of difficulty, discrimination, as well as analyzing the factors that cause the question items to be difficult and have poor discrimination. The respondents in this research were 200 students taken from three junior high schools in DIY. Meanwhile, qualitative data analysis is used to explain the difficulty index and discrimination of the test items. The results of this research show that the Aiken value is valid, and the reliability estimates are quite good. The results of analysis based on the CTT and IRT approaches obtained a good mean index of difficulty and discrimination. These findings imply that the validated and reliable question items can be used as a reference for developing higher-quality assessments to improve student learning outcomes.