This study aims to describe and analyze the effectiveness of the mufrodat learning program to improve Arabic language skills in class VIII A students at MTs Jamaluddin Lekong Rebuk. The research design used was a quasi-experimental design in the form of a nonequivalent pre-test-post-test group one-taled group design. The data collection method used purposive sampling technique. To find out the difference (increase) from before and after the action was carried out, a paired sample t test was carried out with the hypothesis H0 = there was no difference between before and after the action was carried out, and H1 = there was a difference between before and after the action. and after the action. The results of the t-test showed that after being given the action, the post-test score of class VIII A students was greater than the pre-test result with the results of tcount(9.996) > ttable(2,093) so that the conclusion was that there was a significant difference in the ability of students before and after the treatment was applied. As for knowing its effectiveness, it is carried out using the n-gain theory, from the results of data analysis using the N-gain theory, the increase is only able to reach 37%. So that the mufrodat-based learning that the researchers did was considered less effective in improving the students' Arabic language skills. This is because the research time is relatively very short, so the results shown are also less than optimal. It is recommended for further research to conduct research with adequate time so that the results achieved are maximal.