<![CDATA[ABSTRACT Character strengthening in elementary schools can be fostered through learning experiences that engage students with social issues closely connected to their everyday lives. This study examines the use of Deep Learning with bullying issues as a context for strengthening the character of fourth-grade students at SD Negeri Suko II. A quantitative approach was employed using a quasi-experimental method with a nonequivalent control group design. The population consisted of 102 students distributed across three classes, while the data analyzed were obtained from 68 students, comprising 33 students in the experimental group and 35 students in the control group. Participants were selected through probability sampling using the simple random sampling technique. Data were collected using a character-strengthening test administered before and after the learning intervention. The experimental group engaged in learning activities involving bullying case analysis, discussion, collaboration, and reflection, whereas the control group received conventional instruction. Score improvement was analyzed using N-Gain, while differences in posttest achievement were examined using the Mann–Whitney U test. The experimental group achieved an N-Gain of 0.76 (76.68%), categorized as high, while the control group obtained 0.43 (43.36%), categorized as moderate. The Mann–Whitney test produced an Asymp. Sig. (2-tailed) value of < 0.001, indicating a statistically significant difference in posttest achievement between the two groups. These findings position Deep Learning based on bullying issues as a contextual approach with potential to support character strengthening among elementary school students. ABSTRAK Penguatan karakter di sekolah dasar dapat ditempatkan dalam pengalaman belajar yang mempertemukan siswa dengan persoalan sosial yang dekat dengan kehidupan mereka. Penelitian ini menelaah penggunaan pembelajaran Deep Learning dengan isu bullying sebagai konteks untuk memperkuat karakter siswa kelas IV di SD Negeri Suko II. Penelitian menggunakan pendekatan kuantitatif dengan rancangan quasi experimental nonequivalent control group. Populasi terdiri atas 102 siswa dalam tiga kelas, sedangkan data penelitian dianalisis dari 68 siswa yang terdiri atas 33 siswa kelompok eksperimen dan 35 siswa kelompok kontrol. Pemilihan peserta dilakukan melalui probability sampling dengan teknik simple random sampling. Data dikumpulkan melalui tes penguatan karakter yang diberikan sebelum dan sesudah pembelajaran. Kelompok eksperimen mengikuti pembelajaran melalui analisis kasus bullying, diskusi, kolaborasi, dan refleksi, sedangkan kelompok kontrol mengikuti pembelajaran konvensional. Peningkatan skor dianalisis menggunakan N-Gain, sementara perbedaan capaian posttest diuji dengan Mann–Whitney U. Kelompok eksperimen memperoleh N-Gain sebesar 0,76 (76,68%) dalam kategori tinggi, sedangkan kelompok kontrol sebesar 0,43 (43,36%) dalam kategori sedang. Uji Mann–Whitney menghasilkan Asymp. Sig. (2-tailed) < 0,001, yang menunjukkan adanya perbedaan signifikan pada capaian posttest kedua kelompok. Temuan ini menempatkan pembelajaran Deep Learning berbasis isu bullying sebagai pendekatan kontekstual yang berpotensi mendukung penguatan karakter siswa sekolah dasar.]]>
Copyrights © 2026
