<![CDATA[Verbal bullying in secondary school settings poses significant negative effects on students' psychological well-being and requires systematic intervention. This study aimed to examine the effectiveness of group guidance services using the role-playing technique in reducing verbal bullying behavior among students at SMK Negeri 8 Palu. A quantitative experimental approach with a one-group pretest-posttest design was employed. Ten students were selected through purposive sampling based on school counselor identification. Data were collected using a verbal bullying questionnaire (r = 0.532; α = 0.950) administered before and after three sessions of group guidance. Shapiro-Wilk normality tests confirmed normal data distribution (pretest: sig. = 0.315; posttest: sig. = 0.624; p > 0.05). Paired Samples t-test results indicated a statistically significant difference (t = 21.177; Sig. 2-tailed = 0.000; p < 0.05), with a Cohen's d effect size of 6.59, classified as very large. The mean verbal bullying score decreased from 70.40 to 61.90, with 9 of 10 students shifting from the high to moderate category. The role-playing technique within group guidance services proved both statistically and practically effective in reducing students' verbal bullying behavior. Future research is recommended to incorporate control groups, larger samples, and multiple data collection methods for more comprehensive findings. ABSTRAK Perilaku bullying verbal di lingkungan sekolah menengah berdampak negatif terhadap kondisi psikologis siswa dan memerlukan intervensi yang sistematis. Penelitian ini bertujuan menguji efektivitas layanan bimbingan kelompok dengan teknik role playing dalam mengurangi perilaku bullying verbal siswa SMK Negeri 8 Palu. Penelitian menggunakan pendekatan kuantitatif eksperimen dengan desain one group pretest-posttest. Sampel berjumlah 10 siswa yang dipilih melalui purposive sampling berdasarkan hasil identifikasi guru BK. Data dikumpulkan menggunakan angket perilaku bullying verbal (r = 0,532; α = 0,950) yang diadministrasikan sebelum dan sesudah tiga sesi layanan bimbingan kelompok. Hasil uji normalitas Shapiro-Wilk menunjukkan distribusi data normal (pretest: sig. = 0,315; posttest: sig. = 0,624; p > 0,05). Uji Paired Samples t-test menghasilkan nilai t = 21,177 dengan Sig. (2-tailed) = 0,000 (p < 0,05), disertai nilai effect size Cohen's d = 6,59 yang termasuk kategori sangat besar. Rata-rata skor bullying verbal menurun dari 70,40 (pretest) menjadi 61,90 (posttest), dengan 9 dari 10 siswa berpindah dari kategori tinggi ke sedang. Teknik role playing dalam layanan bimbingan kelompok terbukti efektif secara statistik maupun praktis dalam mengurangi perilaku bullying verbal siswa. Penelitian selanjutnya disarankan menggunakan kelompok kontrol, sampel yang lebih besar, serta instrumen pengumpulan data yang lebih beragam.]]>
