This study aimed to describe the application of card sort strategy for learning outcomes on the material of temperature and their changes. Some things that were described in the study include the feasibility of learning, learning outcomes and student response. The learning outcomes had been measured include attitude competency, knowledge competency and skills competency. The method that used was quasi-experimental design in the form of nonequivalent control group. The samples that used were students of VII A class as an experimental class and VII G class as a control class. The results showed that the feasibility study on the experimental class of 3.50 and 3.51 for the control class. The average feasibility study on two class achieve very good performing category. In the attitude competency showed that the experimental class and control class get the same value that is 4 (four) with the criteria very well. In the knowledge competency, the average knowledge competency of experimental class is higher than the control class. Evidenced from the average score of posttest experimental class is 2.99 and control class is 2.69. Results of the analysis showed that t = 2.4 and ttable = 1.67 with a significance level α = 0.05. So t> ttable, then Ho is rejected and H1 accepted that on average the experimental class better than the control class. In the skills competency showed that the experimental class is higher than the control class. At the first meeting, the average yield of skills competency the experimental class 3.15 and the control class reached 3.11. At the second meeting of the experimental class reached 3.14 and control class reached 3.09. In the results of student responses indicated that the application of the card sort strategy to get a response by 78% is included in the strong criteria. Keywords: card sort strategy, learning outcomes, response.