<![CDATA[Scientific literacy at Ngedukelu Elementary School is still relatively low. This low level of scientific literacy is influenced by various factors, including limited access to age-appropriate reading materials, a lack of content relevant to the local cultural context, and a learning approach that does not fully accommodate the characteristics and needs of students in this area. This study aims to determine the improvement of scientific literacy through the implementation of child-friendly reading materials based on ethnoscience. The type of research used was Classroom Action Research (CAR) with the Kemmis and Taggart model, which consists of four stages: planning, action implementation, observation, and reflection. Data analysis was carried out using a scientific literacy test using indicators adapted from Khouidin et al. (2017). Data analysis was conducted quantitatively according to the characteristics of CAR. The results of the first cycle test showed that 3 students (23.08%) were in the low category, 7 students (53.85%) were in the medium category, and 3 students (23.08%) were in the high category. As a class, 10 students achieved a score of 70 or higher, with a completion percentage of 76.93%. The results of the science literacy test in Cycle II showed a more optimal improvement. Of the 13 students, only 1 (7.69%) was in the low category, 5 (38.46%) were in the medium category, and 7 (53.85%) were in the high category. The number of students achieving a score of 70 or higher increased to 12, with a classical completion percentage of 92.31%. Thus, it can be concluded that implementing child-friendly reading materials based on Ngada ethnoscience has been proven effective in improving the science literacy skills of fourth-grade students at Ngedukelu Elementary School.]]>
