Tujuan dari penelitian ini adalah untuk mengetahui apakah ada perbedaan yang signifikan pada siswa sebelum dan sesudah menggunakan flower writing strategy sebagai strategi dalam pengajaran menulis pada siswa kelas XI MAN Tarakan. Penelitian ini menggunakan rancangan eksperimental dan siswa kelas sepuluh di MAN Tarakan sebagai populasi. Jumlah sampel dalam penelitian ini adalah 40 siswa yang dipilih dengan teknik purposive sampling. X MIA I menjadi satu-satunya kelas yang menjadi contohnya. Terdiri dari 40 siswa. Ada tiga tahap dalam penelitian ini yaitu pretest, treatment dan posttest. Dalam pretest, para siswa diminta menulis teks deskriptif yang berhubungan dengan teman, keluarga, atau orang terkenal. Setelah pretest, peneliti memberi treatment. Kelas treatment diajar dengan menggunakan flower writing strategy sebagai strategi. Setelah memberikan treatment, siswa diminta kembali menulis teks deskriptif yang berhubungan dengan teman, orang terkenal, atau keluarga sebagai posttest. Hasil penelitian menunjukkan bahwa ada perbedaan skor rata - rata antara pretest dan posttest. Hal itu dibuktikan dengan hasil skor rata-rata siswa. Skor mereka adalah 55,35 di pretest dan 67,55 di posttest. Artinya, peningkatan nilai rata-rata siswa adalah 12,20. Dapat dilihat bahwa peningkatan nilai rata-rata siswa pada posttest lebih tinggi daripada pretest dan menunjukkan ada perbedaan yang signifikan dari nilai tulisan siswa antara pretest dan posttest. Selain itu, hasil dari paired sample test menunjukkan bahwa -8,008 lebih tinggi dari t tabel (2,023), artinya tulisan siswa mengalami perbaikan setelah memberikan treatment. Sedangkan hasil sig (2 tailed) 0,000 lebih rendah dari 0,05, artinya flower writing strategy. Dari kedua hasil tersebut dapat disimpulkan bahwa terdapat perbedaan yang signifikan antara pretest dan posttest dalam penulisan teks deskriptif setelah treatment yang diberikan oleh peneliti dan flower writing strategy sebagai strategi efektif dalam pengajaran menulis teks deskriptif.

The objective of this research was to find out whether there was any significant difference on students before and after using flower writing strategy as the strategy in teaching writing at the tenth grade students of MAN Tarakan. This research used pre experimental design and tenth grade students at MAN Tarakan as the populations. The numbers of sample in this research were 40 students which were chosen by using purposive sampling technique. X MIA become the only class that was being as the example. It consisted of 40 students. There were three steps in this research namely pretest, treatment and posttest. In pretest, the students were asked to write a descriptive text related to friend, famous person, or family. After pretest, the researcher gave treatment. The treatment class was taught by using flower writing strategy as strategy. After giving the treatment, students were asked to write descriptive text related to friend, famous person, or family again in the posttest. The research result showed that there was different mean score between pretest and posttest. It was proved by the result of the students’ mean score. They were 55.35 in pretest and 67.55 in the posttest. It meant that the improvement of the students’ mean score was 12.20. It can be seen that the improvement of the students’ mean score in the posttest was higher than pretest and it indicated that there was significant difference from students’ writing score between pretest and posttest. In addition, the result of paired sample test showed that -8.008 higher than t table (2.023), it means that the students writing had an improvement after giving the treatment. Meanwhile, the result of sig (2 tailed) 0.000 was lower than 0.05, it meant that the flower writing strategy can give any contribution to students’ writing skill in descriptive text. From both of the result it can be concluded that there was significant different between pretest and posttest in writing descriptive text after the treatment that was given by the researcher and flower writing strategy as strategy was effective in teaching writing descriptive text.