This study aims to determine science learning by using experimental methods on cognitive learningoutcomes of fifth grade students of SD Negeri 195 Palembang. This research is a quantitativeresearch using the experimental method of True-Experimental Design research in the form ofPosttest-Only Control Design. The research population was all fifth grade students of SD Negeri 195 Palembang, totaling 131 students. The sampling technique used is Probability Sampling type Simple Random Sampling. The sample in this study was taken from class VA which amounted to 38 students and class VB which amounted to 32 students. Data collection techniques using tests ofstudent cognitive learning outcomes and observation. The instrument for measuring cognitive learning outcomes uses a posttest in the form of an objective test description. The validity of the test instrument was tested using the Product Moment correlation formula and to determine the reliability, it was tested using the Cronbach Alpha formula. Data analysis technique using t test. The results showed that the experimental class group obtained an average posttest value of 83.16 and the control class obtained an average posttest score of 57.97. The results of the t-test are 6.432 with a t-table of 1.996, meaning that it can be concluded that there is a significant effect between learning science using experimental methods on cognitive learning outcomes of elementary school students.