Penelitian ini bertujuan untuk mengetahui keefektifan penggunaan teknik CIRC dalam membaca teks narrative bagi siswa kelas sembilan SMP Muhammadiyah 1 Tarakan. Metode penelitian ini adalah quasi experimental. Sampel diambil dengan menggunakan random sampling. Ada dua kelas yang diambil sebagai sampel dalam penelitian ini. Kelas eksperimen diajarkan menggunakan teknik CIRC sedangkan kelas kontrol diajarkan menggunakan metode langsung. Instrumen penelitian ini adalah tes pemahaman membaca. Data yang dianalisis adalah data nilai hasil pretest dan posttest dari kelas ekperimen dan kelas kontrol. Dalam menganalisis data, peneliti menggunakan sampel independent T -Test dengan SPSS 16.0. Hasil dari penelitian menunjukkan bahwa nilai rata-rata kelas eksperimen adalah 62,40 dan nilai rata-rata kelas control adalah 54,80. Hasil dari perhitungan sample independent t-test formulasi di spss 16.0 dari perbandingan kelas experiment dan kelas control ialah 1.153 Dengan kata lain, hipotesis nol (H0) ditolak dan hipotesis alternatif (Ha) diterima. Hasil analisis data membuktikan bahwa hasil pengajaran menggunakan CIRC technique lebih tinggi dibandingkan tanpa menggunakan CIRC technique. Peneliti menyimpulkan bahwa siswa yang diajarkan menggunakan CIRC technique efektif daripada siswa yang diajar tanpa menggunakan CIRC technique di kelas IX SMP Muhammadiyah 1 Tarakan.

This study aimed at finding out the effectiveness of using CIRC technique in teaching writing descriptive text for ninth grade students of SMP Muhammadiyah 1 Tarakan. The method of the study was quasi experimental. The samples were taken by using random sampling. There were two classes as the samples of this research. The experimental class was taught by using CIRC technique and control class was taught using direct method. The instrument of the study was reading comprehension test. The analyzed data was from the score data pretest and posttest of the experimental class and control class. In analyzing the data, the researcher used independent sample T-Test with SPSS 16.0. The results of the study showed that the mean score of the experimental class was 62.40 and the mean score of the control class was 54.80. The result of computation by using independent sample t- test formulation in SPSS 16.0 from equal variance between class experiment and control was 1.153. In other words, the null hypothesis (H0) was rejected and the alternative hypothesis (Ha) was accepted. The results of the analysis data results prove that teaching using clustering technique of is higher than without using a clustering technique. The researcher concluded that students who were taught by using CIRC technique was more effective than the students who were taught without using CIRC technique at the ninth grade of SMP Muhammadiyah 1 Tarakan.