This study aims to determine the difference in science learning outcomes between students who study with the Teams Games Tournament (TGT) learning model assisted by hidden chart media. The students class IV of MI Al Washliyah Perbutulan. This type of research is a quasi-experimental research (quasi-experiment). The population in this study were fourth grade students at MI Al Washliyah Perbutulan campus 2. The population in this study was 60 students. The sampling technique used in this study is the Stratified Random Sampling technique. The sample selected in this study was class IV D. From these data, the highest score, the lowest score, and the average value and mastery of learning at MI Al Washliyah Perbutulan increased in cycle 2. The highest value in the initial data was 85, cycle 1 is 95, increased in cycle 2 to 100, the lowest value in the initial data is 40, cycle 1 is 60 increased in cycle 2 becomes 65, the average value in the initial data is 60.5, cycle 1 is 76.3 increased in cycle 2 becomes 88.7 and learning completeness on the initial data is 25%, cycle 1 is 50% increased in cycle 2 to 80%. In cycle 2, almost all of the children's scores increased above the KKM, which is 75. This proves that the TGT learning model assisted by the hidden chart media has a significant effect on student learning outcomes in science learning compared to conventional learning models.