<![CDATA[The general objectives of this research are: (1) cooperative learning model is more effective to use between TGT and NHT, (2) the effect learning activity to student’s achievement, (3) there is an interaction between learning models and learning activity to learning achievement of mathematics. This is a quasi-experimental research. The research populations are the seven grade of SMP 9 Yogyakarta in Academic Year 2013/2014. Samples are taken with cluster random sampling technique obtained experimental class I and experimental class II. The data were obtained through test and questionnaire. The test instrument used test of item validity and test of reliability. In the analysis of data, the researcher used two-way analysis of variance test for using requirements analysis are normality and homogeneity tests. The results show that the value of FA = 24.364 > Ftable = 4.00; FB = 209.683 > Ftable = 3.15; and FAB = 0.664 < Ftable = 3.15 which means that all of the hypotheses are accepted. The result of the study showed TGT is more effective to be used than NHT, high learning activities give the best achievement of mathematics means that learning activities will give an effect to the students’ achievement, there is no interaction between learning model and learning activity on the learning achievement of mathematics means that on each learning activity levels (high, medium, low) mathematics achievement of TGT is always better than NHT, as well as on each learning model (TGT, NHT) mathematics achievement of high learning activities is always better than mathematics achievement of medium and low learning activity.]]>
