This article describes students’ learning activities during online-based learning by applying Think Pair Share and Student Team Achievement Division cooperative learning models. Type of the research in this article is a quasi-experiment with a non-equivalent group design. The population and sample involved fourth grade students of SDN Simomulyo I Sukomanunggal Surabaya, students at grades IV-G and IV-H. It was the method effectiveness test, and the data were analyzed by using the Kolmogorov Smirnov test. The students’ learning activities data were normally distributed and homogeneous by using the t-test. If it was not normal and homogeneous, it used the Mann-Whitney U-Test. To analyze descriptively, it was conducted by analyzing the results of the observation sheet to find out the percentage value of students’ activities. The results showed that (1) the students’ learning activities in class IV-G applied Think Pair Share cooperative learning model at SDN Simomulyo I, the lowest score was 37 or 72.5% (medium), the highest was 51 or 100% ( high), the rest are above 80% (high), and the average score of activity was 46.27. (2) Students’ learning activities in class IV-H applied Student Team Achievement Division cooperative learning model at SDN Simomulyo I, the lowest score was 39 or 76.4% (medium), the highest was 51 or 100% (high), the rest was above 80% (high), and the average score of activeness was 46.00. (3) There was no difference between the students’ learning activities in class IV-G by applying the Think Pair Share cooperative learning model and class IV-H by applying the Student Team Achievement Division cooperative learning model at SDN Simomulyo I Surabaya with a significance level of 0.791 > 0, 05.