The purpose of this study was to determine the effectiveness of blended learning on the mathematics learning outcomes of students at SMPN 2 Kota Bima. The sample in this study was class VII7 which was taught using the blended learning method and class VII6 using the conventional method. This research is a quasi-experimental research with a pre-test post-test design model, namely research that provides test questions before and after learning using blended learning methods and conventional methods. The research instruments used in this study were pre-test and post-test questions in the form of descriptions of the numbers material. The data collection technique was to use the technique of documenting the results of the pre-test and post-test given to both the experimental and control classes. The data analysis technique used sample inferential analysis technique using the Normalized Gain (N-Gain) test through the SPSS 20.0 for Windows program by first testing for normality and homogeneity. based on the results of the hypothesis testing carried out by the through-NGain test, it was found that the average N-Gain value of the Control class with a 95% confidence level was 60.17% with a minimum N-Gain score of 44.44 and an N-Gain with a maximum score of 76.19. Based on these results it can be concluded that learning using conventional methods for class VII students of SMPN 2 Kota Bima is in the category of being quite effective for improving learning outcomes. For the Experiment class, the average N-Gain score with a confidence level of 95% was 43.74% with a minimum N-Gain score of 21.43 and a maximum N-Gain score of 72.22. Based on these results it can be concluded that mathematics learning using the blended method for class VII students of SMPN 2 Kota Bima is in the less effective category for improving learning outcomes.