<![CDATA[This study aims to: (1) analyze the effectiveness of the Concrete-Image-Abstract (KGA) approach in improving the mathematical reasoning and proof process of second-grade students of SDN 001 Bentiar Besar; and (2) measure the magnitude of the increase in mathematical reasoning through KGA compared to conventional learning. A quasi-experimental Non-Equivalent Control Group design was applied with 60 students (n = 30 per class). Data were collected using a validated mathematical reasoning and proof test (10 items, maximum score 100) and analyzed through the Shapiro-Wilk, Levene, independent t-test, N-Gain, and Cohen's tests. The results showed: (1) there was a significant difference in post-test scores between the experimental and control classes (t = 9.72; p < 0.001); (2) the N-Gain of the experimental class g = 0.62 (medium-high category) compared to the control g = 0.26 (low category); and (3) large effect size (d = 1.47). The conclusion of the study confirms that the KGA approach which integrates concrete manipulatives, pictorial representations, and symbolic abstractions substantially and fairly improves the mathematical reasoning and proof of lower grade students in elementary schools in remote areas. ABSTRAK Penelitian ini bertujuan: (1) menganalisis efektivitas pendekatan Konkrit-Gambar-Abstrak (KGA) dalam meningkatkan proses penalaran dan pembuktian matematis siswa kelas II SDN 001 Bentiar Besar; dan (2) mengukur besaran peningkatan penalaran matematis melalui KGA dibandingkan pembelajaran konvensional. Desain kuasi-eksperimen Non-Equivalent Control Group diterapkan dengan 60 siswa (n = 30 per kelas). Data dikumpulkan menggunakan tes penalaran dan pembuktian matematis tervalidasi (10 butir, skor maksimum 100) dan dianalisis melalui uji Shapiro-Wilk, Levene, uji-t independen, N-Gain, dan Cohen's. Hasil penelitian menunjukkan: (1) terdapat perbedaan signifikan skor postes antara kelas eksperimen dan kontrol (t = 9,72; p < 0,001); (2) N-Gain kelas eksperimen g = 0,62 (kategori sedang-tinggi) dibandingkan kontrol g = 0,26 (kategori rendah); dan (3) effect size besar (d = 1,47). Simpulan penelitian menegaskan bahwa pendekatan KGA yang mengintegrasikan manipulatif konkret, representasi piktoral, dan abstraksi simbolik secara substansial dan berkeadilan meningkatkan penalaran dan pembuktian matematis siswa kelas rendah di sekolah dasar daerah terpencil. ]]>
Copyrights © 2026
