<![CDATA[The problem in this study was the low emotional social ability of children. The study aimed to determine the influence of Numbered Heads Together Learning model towards the social-emotional ability of children aged 5-6 years at the Assisi Kindergarten Medan Academic Year 2017/2018. This research is the experimental research of Posttest-Only Control Group Design. The population of this research was the entire group B in the Assisi Kindergarten of Medan which was 110 children. Sampling techniques were carried out with random sampling, i.e. selecting samples randomly with a raffle. The first take is to select the experiment class and the control class with the draw 2 of the 5 classes. The first step was taking a random class, which was an experimental class and the second step was taking a random class for the control class. For the experimental class, 25 children were taught by using Numbered Heads Together learning model and in the control class with 25 children was taught by learning demonstration model. Based on the results of the study and the discussion, it’s found that the children of the experiment class has the lowest value 6, the highest value 12 and the average value of 10.4 higher than that of the control class with the lowest value 4 and the highest value 11 with the average of 7.84. This showed that the children's emotional social ability in the experimental class is better than the child of the control class. From the calculation result obtained value T_count = 4.76; this value is compared to the Ttable with value (DK = 〖 (n 〗 _ 1 + n_2) – 2 = 48 and α = 0.05). The value of T table with α = 0.05 is obtained 1.694. So it can be deduced T_count ˃ T_table or 4.76 ˃ 1.694. Thus the H_0 hypothesis was rejected and H_a was accepted, so it could be stated "There is a significant influence learning model Numbered Heads Together to the emotional social ability of children aged 5-6 years in Assisi Kindergarten Medan.]]>
