This research aims to determine if there are differences in the teaching methods in statistical subjects to the learning outcomes if IQ and student motivation are controlled. With a population of all students of SMPI Al A'la, the samples in this study were as much as 60 students divided into 4 classes containing 15 students each. Treatment is given to teachers in D-class, namely administering the Collaboration Problem Solving learning method. The Dependent variables in this study are the outcomes of student learning, while the IQ and the learning motivation are used as covariate variables. The results showed that there is a significant difference in student learning outcomes given different learning methods if controlled IQ and motivation for learning. By conducting advanced tests with Tukey HSD method, it can also be known that the learning results obtained in class with the learning method of Collaboration Problem Solving have a significant difference compared to class A using method Learning lecture with a value of sig. 0.14, class B uses an individual assignment method with a value of sig. 0.005, and class C uses the method of learning discussions with the value of SIG. 0.004. If you see a difference in reality, then it can be known that the Collaboration Problem Solving learning method is applied to the D-class, providing the best learning outcomes compared to other methods.