<![CDATA[Abstrak Penelitian ini bertujuan untuk mengetahui perbedaan hasil belajar siswa antara model pembelajaran think talk write dan pembelajaran ekspositori di Kelas VII SMP Negeri 15 Kota Bengkulu. Populasi pada penelitian ini adalah siswa kelas VII SMP Negeri 15Kota Bengkulu tahun ajaran 2016/2017. Sampel diambil dengan teknik purposive samplingsehingga diperolehkelas VII A sebagai kelas eksperimen dan VII B sebagai kelas kontrol. Teknik pengumpulan data menggunakan tes. Instrumen penelitian berupa posttest berbentuk soal uraian. Instrumen posttest di uji cobakan di kelas uji coba, lalu dilakukan analisis dengan uji validitas (empiris), uji reliabilitas, uji tingkat kesukaran dan uji daya pembeda soal.Hasil posttest menunjukkan kelas kontrol dengan jumlah siswa 33 orang, memperoleh rata-rata hasil belajar siswa 60,70 dan varians 119,04. Sedangkan kelas eksperimen dengan jumlah siswa 32 orang, memperoleh rata-rata hasil belajar siswa 75,88 dan varians 70,71Teknik analisis data yang digunakan adalah uji prasyarat analisis yang terdiri dari uji normalitas dan uji homogenitas. Kemudian pengujian hipotesis menggunakan uji-t satu pihak. Berdasarkan perhitungan uji-t satu pihak dengan taraf signifikan 5% dan derajat kebebasan 63 diperoleh maka dapat disimpulkan bahwa terdapat perbedaan yang signifikan antara rata-rata hasil belajar siswa dengan menggunakan model pembelajaran think talk write dan pembelajaran ekspositori di kelas VII SMP Negeri 15 Kota Bengkulu. Kata Kunci:hasil belajar, think talk write, ekspositori AbstractThe purpose of this study was to investigate the differences of learning outcomes ability between students given model of learning ThinkTalk-Write with Expository in first grade of SMP Negeri 15 Bengkulu city. The population was first grade of SMP Negeri 15 Bengkulu city in the academic year 2016/2017. Then by using purposive sampling technique, VII A was selected as the experimental class and VII B as the control class. Data collecting technique was test. The instrument was essay posttest. The posttest instrument had tested in a trial class, then analyzed its validity, reliability, level of difficulty and discrimination power. The results showed that 33 students of control class got the scores about 60,70 with variances 119,04 while 32 students of experimental class got 75,88 with variances 93,08. Data analysis was performed by precondition testing analysis which consisted of normality and homogeneity testing. Then, the hypothesis was tested by using one side t-test. Based on the calculation of two side t-test with 5% of significant level and 63 of independent degree resulted tcount = 6,269> ttable = 1.998 then it was concluded that in the first grade of SMP Negeri 15 Bengkulu city, there is a signignificant difference between learning outcomes of students who were given model of learning Think-Talk-Write and than the students who were given model of learning Expository. Keyword :learning outcomes. think talk write, expository]]>
