Journal article
Jurnal Khazanah Sekolah Dasar
• August 2017

This research is motivated by fact about the ability of students in solving low math problems. This research aims to; 1) Proving how big problem solving ability of mathematical student improve by using model of problem based learning, 2) proving how big problem solving ability of mathematical student improve using conventional learning, 3) Proving difference of problem solving ability of mathematical student using problem based learning model with conventional. The method used is quasi experimental and research design used is nonequivalent Control Group Design. The population of this research is the third grade students of Elementary School in Cileunyi District. The sample of this research is the students of the third grade B Elementary School Sukahati 02 who are in cluster 5 and the third grade B Elementary School Cimekar which is in cluster 3. The experiment result of class study which is using problem-based learning model pretest average value is 42,93 and the mean value Posttest 73.2 with an average gain of 0.53 medium criterion. In the control class which is using conventional learning the average pretest score is 41.7 and the mean posttest score is 60.73 with an average gain in 0.32 medium criterion. The result of a difference test of mean obtained 0,000 significance less than 0,05, then rejected. Thus, it can be concluded that there is a difference in problem solving abilities of mathematical students using problem-based learning model with conventional. This research is motivated by fact about the ability of students in solving low math problems. This research aims to; 1) Proving how big problem solving ability of mathematical student improve by using model of problem based learning, 2) proving how big problem solving ability of mathematical student improve using conventional learning, 3) Proving difference of problem solving ability of mathematical student using problem based learning model with conventional. The method used is quasi experimental and research design used is nonequivalent Control Group Design. The population of this research is the third grade students of Elementary School in Cileunyi District. The sample of this research is the students of the third grade B Elementary School Sukahati 02 who are in cluster 5 and the third grade B Elementary School Cimekar which is in cluster 3. The experiment result of class study which is using problem-based learning model pretest average value is 42,93 and the mean value Posttest 73.2 with an average gain of 0.53 medium criterion. In the control class which is using conventional learning the average pretest score is 41.7 and the mean posttest score is 60.73 with an average gain in 0.32 medium criterion. The result of a difference test of mean obtained 0,000 significance less than 0,05, then rejected. Thus, it can be concluded that there is a difference in problem solving abilities of mathematical students using problem-based learning model with conventional.