Abstract: This study aims to determine whether there is an average difference between student learning outcomes are subject to a bamboo dance learning model and think phare share model with interactive multimedia and conventional models to determine which is better. Data collection methods used consisted of test methods and documentation. Preliminary data such as test scores were analyzed using normality test, homogeneity test and one-way ANOVA test so that the experimental groups 1, 2 experimental and control groups can be used as a sample. Final data in the form of evaluation tests analyzed using normality test, homogeneity, classical completeness of test, one-way ANOVA test and t test and the simple linear regression test. The results of the final analysis obtained Fhitung = 53.20 and F-table = 3.24, so H₀ is rejected which means that there are differences in the average learning outcomes of all three groups. In a test of hypothesis 3 is obtained t = 2.83 and table = 1.65, so that H₀ is rejected means experimental group 1 is better than the control class. In the 4th hypothesis test obtained t = 2.96 and table = 1.65 so that H₀ is rejected means that the experimental group 2 was better than the control group. In a hypothesis test five experimental class 1 is R² = 18.91% and in experimental class 2 is R² = 18.145% which means there is a contribution to the liveliness of student learning outcomes in experimental class 1 and 2. the conclusion of this study is the learning outcomes of students who get the bamboo dance and learning model TPS-aided interactive multimedia better than conventional learning model.