This study aims to analyze the literacy skills of fifth-grade elementary school students in science subjects with homogeneous and heterogeneous substances using the RADEC model (Read, Answer, Discuss, Explain, and Create). Experimental research with this type of quasi-experimental design is a quantitative study with two groups: the experimental and control classes. In the experimental class, there were 23 students in the sample, and in the control class, there were 18 students. Data sources used in this study was 16 multiple-choice questions on scientific literacy in science learning. Data were obtained from scientific literacy tests; then, analyzed using the Winstep application with Rasch modeling to obtain information about changes in scientific literacy skills in science learning. The research results were processed using the Winstep Rasch model using stacking analysis in the experimental and control classes. The experimental class, stacking analysis results, provide information that it has increased and is grouped based on the increase in logit value into the very good category 87%, 9% good, and 4% sufficient. In the control class, the logit value was very good, 61% and 39% sufficient. Based on research that has been conducted, it can be concluded that the radec learning model can improve the scientific literacy skills of grade 5 students in science learning.
Read full abstract