This research was an experimental research. The purposes of this research were assessing the effect of Problem Based Instruction model to the problem solving skills and describing student learning activities which were using Problem Based Instruction model. Experimental design in this research was the posttest only control group with two classes of samples, that are experiment class and control class. Problem solving skill score was obtained through the problem solving skill test, and student learning activities score was collected through observation and worksheet documentation during learning process. The test results of an experimental class and a control class were statistically compared by independent sample t-test. The significance (1-tailed) value that obtained was 0.000. It meant that problem solving skill score of experiment class student was better than control class student. It can be concluded that Problem Based Instruction model gave significant effect to the problem solving skill. The average score of student learning activities during the learning process was 56,61. The student learning activities criteria was active enough.