<![CDATA[This study aims to see whether the achievement and improvement of students' mathematical understanding skills who received NHT cooperative learning were better and higher than students who received expository learning and to see how much influence the NHT cooperative learning model had on the students' understanding abilities of the experimental class. . This study uses a quantitative approach with the method used is an experiment. The population in this study were seventh grade junior high school students. The sampling technique used is total sampling. The research design used was a quasi-experimental design, the nonequivalent pretest-posttest control group design. The instruments used were test instruments and unstructured interviews. The data obtained will be analyzed by statistical tests in the form of normality test, homogeneity test, t-test, Mann-Whitney test, n-gain test and effect size. The results of the posttest data analysis showed the value of sig. of 0.005 and the n-gain data shows the value of sig. of 0.00 then Ho is accepted and Ha is rejected. This means that the achievement and improvement of students' mathematical understanding skills using the NHT learning model is better and higher than students using expository learning. The large influence of the NHT type cooperative learning model on the mathematical understanding ability of experimental class students obtained through the effect size test is 0.89 with the criteria classified as large.]]>
