Abstract

Permasalahan dalam penelitian ini adalah rendahnya kemampuan pemecahan matematika siswa kelas V disebabkan karena penerapan model pembelajaran yang kurang melibatkan keaktifan siswa. Penelitian ini bertujuan untuk menguji keefektifan PBL dan TPS terhadap kemampuan pemecahan masalah matematika siswa kelas V. Penelitian ini merupakan penelitian eksperimen quasi dengan bentuk desain quasi Nonequivalent Control Group Design. Populasi penelitian ini adalah siswa kelas V SDN Gugus Kyai Umpak Padureso, Kebumen. Hasil uji hipotesis (1) ketuntasan belajar kelas eksperimen I Zhitung = 3,6055, kelas eksperimen II Zhitung = 2,2873 sedangkan di kelas kontrol Zhitung = -0,0120 menunjukkan bahwa ketuntasan belajar kelas eksperimen I lebih tinggi dari kelas eksperimen II dan kelas kontrol; (2) uji perbedaan dua rata-rata menunjukkan bahwa rata-rata kemampuan pemecahan masalah di kelas eksperimen I lebih tinggi dari kelas eksperimen II dan kelas kontrol; (3) n-gain kelas eksperimen I 0,73 lebih tinggi dari kelas eksperimen II 0,65 dan kelas kontrol 0,60. Jadi dapat disimpulkan bahwa model PBL lebih efektif dibandingkan dengan model TPS dan kelas kontrol.


The problem of this research was the mathematical problem solving skills in the 5th grade were still low due to the application of learning models that did not involve student activeness. In the result, the students had difficulty to do the problem solving tasks. The aim of this study was to test the effectiveness of PBL and TPS model on the problem solving in the  5th grade. This research was a quasi-experimental study. The type of this research was Nonequivalent Control Group Design. The subject of this research was the 5th grade students of Kyai Umpak Cluster Elementary School, Padureso, Kebumen. The hypothesis test results were: (1) the learning completion of the experimental class I zcount = 3,6055, the experimental class II zcount = 2,2873 while the control class zcount = -0,0120 showed that the learning completion in the experimental class I was higher than in the experimental class II and the control class; (2) the two-difference test showed that the average problem solving skills in the experimental class I were higher than the experimental class II and the control class; (3) the n-gain of the experimental class I (0.73) was higher than the experimental class II (0.65) and the control class (0.60). So it can be concluded that the PBL model was more effective than the TPS model.