There is a problem that in grade IV SD Aloysius Semarang found students who feel bored and bored because of monotonous methods and unattractive media so that students are not responsible for student learning so that the impact on student learning outcomes is low, so many students have not reached the KKM which is determined by the school is 80. So this study aims to:
1)         To determine the extent to which students' sense of responsibility has a positive effect on learning outcomes in fourth grade students of SD Aloysius Semarang; 2) To find out how far the learning of the Student Teams Achievement Division (STAD) method assisted by YouTube is more effective in improving student learning outcomes than the conventional method for fourth grade students of SD Aloysius Semarang; 3) to find out to what extent learning using the Student Teams Achievement Division (STAD) method can increase the number of students who reach the KKM in fourth grade students of SD Aloysius Semarang. This research applies an experimental quantitative approach. This type of experimental research is referred to as true experiments. The research design used was Prettest-posttest Only Control Group. with research subjects amounted to 57 fourth grade students of SD Aloysius Semarang. This study consisted of two classes: 1) the experimental class applied the STAD learning method; 2) the control class implements the Conventional method. The instrument used was a multiple choice test with 20 questions and a student responsibility observation sheet. The data analysis of this study used the influence test, homogeneity test and comparative test, N-Gain test and descriptive statistics assisted by SPSS version 21. The results showed that there was a positive influence of students' sense of responsibility on learning outcomes, both groups had a significant value >5% then is homogeneous. Because the data is homogeneous, the hypothesis H0 is accepted and H1 is rejected. Then the Independent t test is carried out at t value sig = 0.000 = 0.0% <5%, then H0 is rejected or H1 is accepted. The results of the N-gain test analysis showed a significant difference, namely the N-Gain value of the experimental class was 59.8052 or 59.8% and was included in the Quite Effective category, while the N-Gain value of the control class was 43.1052% or 43.5% included in the less category. effective. After treating the Student Team Achievement Divisions (STAD) method assisted by YouTube in the experimental class, there were 68.9% or 20 of the students who reached the KKM. Thus, it can be concluded that the use of the YouTube-assisted Student Team Achievement Divisions (STAD) method is more effective on learning outcomes.