This research aims to examine the differences in the improvement of mathematical problem-solving abilities of MTs students with an open-ended approach and conventional learning. In this research the researchers applied quasi-experimental method, where the samples in this research are students of class VII MTs in Cianjur district and take the subject of students of MTs Tanwiriyyah class VII I a total of 41 students as experimental class who get the learning treatment using the approach Open-Ended and class VII H totaling 40 students as control class that get conventional learning. In this study the instrument that applied the test instrument of the problem-solving ability of mathematics as much as 4 item. The result of N-Gain data analysis using t` test shows students' mathematical problem-solving ability of the experimental class is better than the control class.
