Statistical reasoning ability is one of the essential skills in developing competence, which is one of the Sustainable Development Goals (SDGs). This study aims to explore the statistical reasoning ability of junior high school students in descriptive statistics learning. The investigation directs students to determine their level of statistical reasoning ability. The Rasch model analysis analyzes whether there is a bias in the item questions used in exploring these abilities based on demographic factors such as gender, initial mathematical ability, and ethnic background. This study uses a mix-method (a combination of quantitative and qualitative methods). Based on the Rasch Model analysis, quantitative and qualitative data were obtained from 22 students utilizing a statistical reasoning ability test instrument based on the ethnomathematics context, which was valid and reliable. The results of data analysis show that students have different levels of statistical reasoning abilities and are related to students' initial mathematical ability. Students with low initial mathematical skills are at level 1 idiosyncratic, while students with high initial mathematical skills are at procedural level 4. The results of the DIF (Differential Item Functioning) analysis showed no significant difference between the use of the ethnomathematics context-based test instrument and the demographic factors of the research subjects. This study concludes that students can explore statistical reasoning ability through ethnomathematics-based problem solving but still pay attention to other demographic factors.