<![CDATA[Mathematics is considered a difficult subject matter for students, which causes problem-solving processes and low learning outcomes. This collaborative classroom action research (PTKK) was carried out to improve the process and results of learning mathematics on the topic of addition and subtraction of 1st grade students of SD Negeri Randuacir 03 Salatiga. It was carried out over two cycles with two meetings in each cycle. The data collection technique uses test techniques in the form of posttests and non-tests in the form of observations, with data collection tools using posttest questions, student learning process observation instruments, and teacher activity observation instruments. The posttest questions were tested for validity with a significance of 0.05 and reliability with the reliable category. The results of the research obtained (1) that the problem-based learning model can improve the learning process in class I, which is known from the results of observations of student learning processes and teacher activities in the pre-cycle; the percentages were 42.85% and 57.14 in the category not good, cycle I 72, 32%, and 71.42% in the good category, and cycle II 85.71% and 92.85% in the very good category; and (2) that the problem-based learning model can improve learning outcomes with a percentage of 78.57% from the pre-cycle to cycle I and 100% from cycle I to cycle II. The conclusion from the research results is that the problem-based learning model can improve the mathematics learning process and outcomes of 1st grade students. Abstrak Matematika dianggap sebagai muatan pelajaran yang sulit bagi siswa yang menyebabkan proses pemecahan masalah dan hasil belajar rendah. Penelitian Tindakan Kelas Kolaboratif (PTKK) ini dilakukan untuk memperbaiki proses dan hasil belajar matematika topik penjumlahan dan pengurangan di kelas I SD Negeri Randuacir 03 Salatiga yang dilaksanakan selama dua siklus dengan tiap siklus dua pertemuan. Teknik pengumpulan data menggunakan teknik tes berupa posttest dan nontes berupa observasi, dengan alat pengumpulan data menggunakan soal posttest, instrumen observasi proses belajar siswa, dan instrumen observasi aktivitas guru. Soal postest diuji validitas dengan signifikansi < 0,05 dan uji reliabilitas dengan kategori reliabel. Hasil penelitian diperoleh (1) model Problem Based Learning dapat meningkatkan proses pembelajaran di kelas I yang diketahui dari hasil observasi proses belajar siswa dan aktivitas guru pada pra siklus diperoleh presentase 42,85% dan 57,14 dengan kategori kurang baik, siklus I 72,32% dan 71,42% dengan kategori baik, dan siklus II 85,71% dan 92,85% dengan kategori sangat baik; (2) model Problem Based Learning dapat meningkatkan hasil belajar dengan presentase 78,57% pada pra siklus ke siklus I dan 100% dari siklus I ke siklus II. Simpulan dari hasil penelitian bahwa model Problem Based Learning dapat meningkatkan proses dan hasil belajar matematika siswa kelas I.]]>
