ABSTRAK
Tujuan penelitian ini adalah untuk mengetahui:(1) manakah yang mempunyaihasilbelajarmatematikalebihbaik, siswa yang dikenaimodel pembelajarankooperatiftipeGIdengansaintifik, model pembelajarankooperatiftipeTPSdengansaintifikataupembelajaranklasikaldengansaintifik. (2) manakah yang mempunyaihasilbelajarmatematika yang lebihbaik, siswadengankemampuanpenalarantinggi, kemampuanpenalaransedangataukemampuanpenalaranrendah. (3)padamasing-masing model pembelajaran, manakah yang mempunyaihasilbelajarlebihbaik, siswa dengan kemampuan penalarantinggi, sedang atau rendah. (4)padamasing-masingkemampuanpenalaran, manakah yang mempunyai hasil belajar matematika lebih baik, siswa yang dikenaimodel pembelajarankooperatiftipeGIdengansaintifik, model pembelajarankooperatiftipeTPSdengansaintifikataupembelajaranklasikaldengansaintifik.
Jenis penelitian yang digunakan dalam penelitian ini adalah eksperimental semu dengan desain faktorial 3 × 3. Populasi dalam penelitian ini adalah seluruh siswa kelas VIII SMP Negeri di Kabupaten Karanganyar. Pengambilan sampel dilakukan dengan stratified cluster random sampling. Sampel dalam penelitian ini berjumlah 280 siswa dengan banyak anggota sampel untuk kelompok eksperimen 1 berjumlah 94 siswa, kelompok eksperimen 2 berjumlah 94 siswa, dan kelompok kontrol berjumlah 92 siswa. Instrumen yang digunakan untuk mengumpulkan data adalah tes hasil belajar matematika dan tes kemampuan penalaran siswa.Uji prasyarat analisis data menggunakan uji Lilliefors untuk uji normalitas sedangkan untuk uji homogenitas menggunakan uji Bartlett, dengan a= 0,05 diperoleh kesimpulan bahwa sampel berasal dari populasi yang berdistribusi normal dan homogen. Uji keseimbangan menggunakan uji Anava satu jalan dengan sel tak sama, dengan a= 0,05 diperoleh kesimpulan bahwa populasi pada kelompok eksperimen dan kelompok kontrol dalam keadaan seimbang. Analisis data menggunakan Anava dua jalan dengan sel tak sama.
Hasilpenelitianini sebagai berikut.(1) Siswa yang dikenai modelpembelajarankooperatiftipeGI dengansaintifikmempunyaihasilbelajarmatematika yang lebihbaikdari siswa yang dikenai model TPS dengansaintifik dan pembelajaranklasikaldengansaintifik. Selain itu, modelpembelajarankooperatiftipeTPS dengansaintifikmempunyaihasilbelajarmatematika yang lebihbaikdaripembelajaranklasikaldengansaintifik. (2) Siswa yang memilikikemampuanpenalarantinggidansedangmempunyaihasilbelajarmatematika yang lebihbaikdibandingkansiswa yang memilikikemampuanpenalaranrendah. Selain itu, siswa yang memilikikemampuanpenalarantinggimempunyaihasilbelajarmatematika yang samadengansiswa yang memilikikemampuanpenalaransedang. (3) Pada model pembelajarankooperatiftipeGI dengansaintifik, siswa yang memilikikemampuanpenalarantinggidan rendah mempunyaihasilbelajarmatematika yang samadengansiswa yang memilikikemampuanpenalaransedang. Selainitu,  siswa yang memilikikemampuanpenalarantinggimempunyaihasilbelajarmatematikalebihbaikdibandingkandengansiswa yang memilikikemampuanpenalaranrendah. Pada model pembelajarankooperatiftipeTPS danpembelajaranklasikaldengansaintifik, siswadengankemampuanpenalarantinggi, sedang, danrendahmempunyaihasilbelajarmatematika yang sama. (4) Padasiswadengankemampuanpenalarantinggi, model pembelajarankooperatiftipeGIdanpembelajaranklasikaldengansaintifikmempunyaihasil yang samadengan model pembelajarankooperatiftipeTPSdengansaintifik. Selainitu, model pembelajarankooperatiftipeGIdengansaintifikmempunyaihasilbelajarlebihbaikdibandingkanpembelajaranklasikaldengansaintifik. Padasiswadengankemampuanpenalaransedangdanrendah, masing-masing model pembelajaranmempunyaihasilbelajar yang sama.
Katakunci:GI dengansaintifik, TPS dengansaintifik, hasilbelajar, kemampuanpenalaran.
ABSTRACT
The objective of this research was to investigate: (1) which learning model of the cooperative learning model of the GI with scientific, the cooperative learning model of the TPS with scientific, or the classical learning with scientific results in a better learning outcomes in mathematics, (2) which students of the students with the high reasoning ability, the students with the moderate reasoning ability, or the students with the low reasoning ability had a better learning outcomes in mathematics, (3) on each learning model, which students of the students with the high reasoning ability, the students with the moderate reasoning ability, or the students with the low reasoning ability had a better learning outcomes in mathematics, and (4) on each reasoning ability, which learning model of the cooperative learning model of the GI with scientific, the cooperative learning model of the TPS with scientific, or the classical learning with scientific results in a better learning outcomes in mathematics.
This research used the quasi experimental research method with the factorial design of 3 x 3. Its population was all of the students in Grade VIII of State Junior Secondary Schools of Karanganyar. The samples of the research were taken by using the stratified cluster random sampling technique. The samples consisted of 280 students. They were divided into three classes, namely: 94 students in Experimental Class 1, 94 students in Experimental Class 2, and 92 students in Control Class. The instruments used to gather the data were test of learning outcomes in mathematics and test of reasoning ability. The pre-requisite tests of the research included the normality test by using Lilliefors test and homogeneity test by using Bartlett test with a = 0.05. The result of the pre-requisite tests show that the the samples of the research were normally distributed and homogenous. The balance test of the research used the one-way analysis of variance with unbalanced cells with a = 0.05. The proposed hypotheses of the research were tested by using the two-way analysis of variance with unbalanced cells.
The results of the research were as follows. 1) The students instructed with the cooperative learning model of the GI with scientific had a better learning outcomes in mathematics than those instructed with the cooperative learning model of the TPS with scientific and those instructed with the classical learning with scientific. Furthermore, the students instructed with the cooperative learning model of the TPS with scientific had a better learning outcomes in mathematics than those instructed with the classical learning model with scientific. 2) the learning outcomes in mathematics of the students with the high and moderate reasoning ability had a better than that of the students with the low reasoning ability. In addition, the students with the high reasoning ability had the same good learning outcomes in mathematics as the students with themoderate reasoning ability. 3) on the cooperative learning model of the GI with scientific, the students with the high and low reasoning ability had the same good learning outcomes in mathematics as the students with the moderate reasoning ability.Moreover, the students with the high reasoning ability had a better learning outcomes in mathematics than those the students with the low reasoning ability. On the cooperative learning model of the TPS with scientific and the classical learning with scientific results in the same good learning outcomes in each of reasoning ability. 4)the students with the high reasoning ability, the cooperative learning model of the GI with scientific and the classical learning with scientific results in the same good learning outcomes in mathematics as the cooperative learning model of the TPS with scientific. Furthermore, the cooperative learning model of the GI with scientific results in a better learning outcomes in mathematics than the classical learning with scientific. The students with the moderate and low reasoning ability had the same good learning outcomes in each of learning model.
Keywords:     GI with scientific, TPS with scientific, learning outcomes, reasoning ability.