PENGEMBANGAN E-LKPD BERBASIS DEEP LEARNING UNTUK MENINGKATKAN KEMAMPUAN BERPIKIR KREATIF SISWA DAN SELF EFFICACY

Fatikhah, Findi Auliya
Ahmad, Ahmad
Lee, Mohd. Faizal Nizam

AKSIOMA: Jurnal Program Studi Pendidikan Matematika
Vol. 15 No. 3 (2026)

Penelitian ini bertujuan mengembangkan E-LKPD berbasis Deep Learning pada materi pola bilangan serta menganalisis efektivitasnya dalam meningkatkan kemampuan berpikir kreatif dan self-efficacy siswa kelas VIII SMP. Penelitian menggunakan metode Research and Development (R&D) dengan model ADDIE yang meliputi tahap Analysis, Design, Development, Implementation, dan Evaluation. Uji efektivitas menggunakan Pretest–Posttest Control Group Design dengan kelas VIII A sebagai kelas eksperimen dan kelas VIII B sebagai kelas kontrol. Instrumen penelitian meliputi lembar validasi ahli, angket self-efficacy, dan tes kemampuan berpikir kreatif. Analisis data dilakukan menggunakan uji validitas, uji normalitas, uji homogenitas, paired sample t-test, independent sample t-test, dan N-Gain. Hasil validasi menunjukkan bahwa E-LKPD memperoleh tingkat kelayakan sebesar 88,75% dengan kategori sangat valid. Hasil analisis menunjukkan bahwa kelas eksperimen mengalami peningkatan yang signifikan pada self-efficacy (p < 0,001) dan kemampuan berpikir kreatif (p < 0,001) dengan nilai N-Gain masing-masing 0,67 (sedang) dan 0,71 (tinggi), sedangkan kelas kontrol memperoleh N-Gain sebesar 0,32 dan 0,35 (kategori sedang). Selain itu, terdapat perbedaan yang signifikan antara kelas eksperimen dan kelas kontrol (p < 0,05). Temuan ini menunjukkan bahwa E-LKPD berbasis Deep Learning berpotensi lebih efektif dalam meningkatkan kemampuan berpikir kreatif dan self-efficacy siswa dibandingkan pembelajaran konvensional. This study aimed to develop a Deep Learning-based Electronic Student Worksheet (E-LKPD) on number patterns and to analyze its effectiveness in improving the creative thinking skills and self-efficacy of eighth-grade junior high school students. The study employed a Research and Development (R&D) approach using the ADDIE model, consisting of the Analysis, Design, Development, Implementation, and Evaluation stages. The product effectiveness was evaluated using a Pretest–Posttest Control Group Design, with Class VIII A serving as the experimental group and Class VIII B as the control group. The research instruments included expert validation sheets, a self-efficacy questionnaire, and a creative thinking skills test. Data were analyzed using product validity analysis, the Shapiro–Wilk normality test, Levene's homogeneity test, paired sample t-test, independent sample t-test, and N-Gain analysis. The validation results indicated that the developed E-LKPD achieved a feasibility score of 88.75%, which was categorized as highly valid. The analysis showed that the experimental group experienced significant improvements in self-efficacy (p < 0.001) and creative thinking skills (p < 0.001), with N-Gain scores of 0.67 (moderate) and 0.71 (high), respectively. In contrast, the control group obtained N-Gain scores of 0.32 and 0.35, both classified as moderate. Furthermore, the independent sample t-test revealed significant differences between the experimental and control groups (p < 0.05). These findings indicate that the Deep Learning-based E-LKPD is more effective than conventional instruction in enhancing students' creative thinking skills and self-efficacy.