The specific objective of this research is to know which are on mathematics learning achievement better, between students who have high intrapersonal intelligence, medium, and low. This study is a quasi-experimental research. The study population was fifth grade students of SD Negeri Kudus District. The sampling technique was conducted stratified cluster random sampling. The sample in this study were students of class V SD 1 Muhammadiyah Kudus and fifth grade students of SD 1 Gondangmanis Kudus. The instrument used to collect data is intrapersonal intelligence student questionnaires. Questionnaires tested before it is used for data retrieval. The validity of questionnaires carried out by the validator, the reliability of the test is tested with the formula KR-20 and reliability of the questionnaire was tested with Alpha formula. Test instruments carried at SD 2 Holy Honggosoco. Analysis of the data used is two-way analysis of variance. Factors used to test the significance of differences in students' intrapersonal intelligence level of learning achievement. Test prerequisite Lillifors Variance Analysis method to test for normality and homogeneity test methods to Barlett. With α = 0.05. Scheffe method 'used for further analysis of variance test if the hypothesis is rejected.