<![CDATA[ABSTRAK  Penelitian ini bertujuan untuk menganalisis (1) perbedaan prestasi belajar fisika danketerampilan berpikir kritis antara kelompok MPBM dan MPK, (2) pengaruh interaksi antara model pembelajaran dan bakat numerik terhadap prestasi belajar fisikadan keterampilan berpikir kritis, (3) perbedaan prestasi belajar fisika dan keterampilan berpikir kritis pada siswa yang memiliki bakat numerik tinggi antara kelompok MPBM dan MPK, (4) perbedaan prestasi belajar fisika dan keterampilan berpikir kritis pada siswa yang memiliki bakat numerik rendah antara kelompok MPBM dan MPK. Penelitian ini merupakan kuasi eksperimen pada siswa kelas X SMA Negeri 2 Singaraja tahun pelajaran 2010/2011.Penelitian ini menggunakan the post-test only non-equivalent control group design.Data yang dikumpulkan dalam penelitian ini adalah prestasi belajar dan keterampilan berpikir kritis siswa. Data prestasi belajar dikumpulkan dengan tes prestasi belajar dan data keterampilan berpikir kritis dikumpulkan dengan tes keterampilan berpikir kritis. Tes prestasi belajar berbentuk pilihan ganda diperluas terdiri dari 20 butir dengan indeks validitas butir berkisar dari r = 0,296 s.d r = 0,832 dan indek reliabelitas alpha Cronbach = 0,836. Tes keterampilan berpikir kritis berbentuk pilihan ganda terdiri dari 25 butir dengan indek validitas butir berkisar dari r = 0,239 s.d r = 0,674 dan indek reliabelitas K-R.20 = 0,625. Data dianalisis secara dekriptif dan menggunakan MANOVA faktorial 2Ã2. Hasil penelitian menunjukkan bahwa. (1) terdapat perbedaan prestasi belajar dan keterampilan berpikir kritis yang signifikan antara kelompok model pembelajaran berbasis masalah dan kelompok model pembelajaran konvensional (F = 15,13; p<0,05),(2) model pembelajaran dan bakat numerik berinteraksi secara signifikan dalam pencapaian prestasi belajar dan keterampilan berpikir kritis (F = 8,67; p<0,05), (3) terdapat perbedaan prestasi belajar dan keterampilan berpikir kritis yang signifikan antara kelompok model pembelajaran berbasis masalah dan kelompok model pembelajaran konvensional padabakat numerik tinggi(F = 19,91; p<0,05), (4) tidak terdapat perbedaan prestasi belajar dan keterampilan berpikir kritis yang signifikan antara kelompok model pembelajaran berbasis masalah dan kelompok model pembelajaran konvensional padabakat numerik rendah (F = 0,83; p>0,05).  Kata kunci : model pembelajaran, bakat numerik, prestasi belajar, dan keterampilan berpikir kritis.  The effect of problem-based learning model toward ââacademic achievement in physics and critical thinking skills based on numerical aptitude  ABSTRACT This research aimed to analyze (1) the differences in physics academic achievement and critical thinking skills among MPBM and MPK groups, (2) the interaction effect between learning models and numerical aptitude to physics academic achievement and critical thinking skills, (3) the differences in physics academic achievement and critical thinking skills between MPBM and MPK groups for students who have a high numerical aptitude, (4) the differences in physics academic achievement and critical thinking skills between MPBM and MPK groups for students who have low numerical aptitude. This research was a quasi-experimental research in class X SMA Negeri 2 Singaraja academic year 2010/2011. This research used post-test only non-equivalent control group design. Data collected in this study were studentâs physics academic achievement and critical thinking skills. Physics academic achievement data was collected by academic achievement test and critical thinking skills was collected by critical thinking skills test. Academic achievement test was extended multiple choices consisting of 20 items with item validity ranged from r = 0.296 to r = 0.832 and Cronbachs alpha index of reliability = 0.836. Critical thinking skills test was multiple choices consisting of 25 items with item validity from r = 0.239 sd = 0.674 and KR.20 index of reliability = 0.625. Data were analyzed using descriptive statistic and MANOVA 2 à 2 factorial. The results found that, (1) there was statistically significant differences in physics academic achievement and critical thinking skills among groups of problem-based learning model and the conventional learning model (F = 15.13, p <0.05), (2) learning models and numerical aptitude interact significantly in determining physics academic achievement and critical thinking skills (F = 8.67, p <0.05), (3) there was statistically significant differences in physics academic achievement and critical thinking skills among groups of problem-based learning model and the conventional learning model for students who have a high numerical aptitude (F = 19.91, p <0.05), (4) there was no statistically significant differences in physics academic achievement and critical thinking skills among groups of problem-based learning model and the conventional learning model for students who have a low numerical aptitude (F = 0.83, p> 0.05).  Key words: problem based learning, numerical aptitude, achievement, and critical thinking skills.]]>
