ABSTRAK
Syahmani. 2018. Pengembangan Model Pembelajaran Konstruktivis i-SMART Menggunakan Pemecahan Masalah Kimia SMA untuk Melatih Keterampilan Metakognisi dan Pemahaman Konsep Siswa. Disertasi, Program Studi S3 Pendidikan Sains, Pascasarjana Universitas Negeri Surabaya. Promotor: Prof. Dr. Suyono, M. Pd. dan Kopromotor: Dr. Z.A. Imam Supardi, M. Si.
Kata-kata kunci: Pembelajaran konstruktivis i-SMART, pemecahan masalah, keterampilan metakognisi, pemahaman konsep

Tujuan penelitian adalah menghasilkan model i-SMART yang valid, praktis, dan efektif sehingga layak untuk meningkatkan keterampilan metakognisi, pemahaman konsep, dan retensi siswa. Penelitian ini dilakukan dalam tiga tahap, yaitu fase pendahuluan, pengembangan, dan pengujian model. Desain uji coba menggunakan pra-eksperimen dengan one group pretest-posttest design. Subyek uji coba terbatas pada 25 siswa SMA Negeri 2 Banjarmasin, sedangkan subjek uji coba luas terdiri 90 siswa dari tiga SMA Negeri di Banjarmasin. Pengumpulan data melalui penilaian validitas pakar, tes, pengamatan, angket, wawancara, dan dokumentasi. Teknik analisis data dilakukan secara deskriptif kuantitatif dan kualitatif, n- gain, uji t berpasangan/uji Wilcoxon, dan uji Kruskal Wallis. Hasil penelitian menunjukkan bahwa: (1) validitas isi model i-SMART ditinjau dari kebutuhan akan pengembangan model dan kebaruan pengetahuan ilmiah dalam kriteria sangat valid, (2) validitas konstruk model i-SMART ditinjau dari saling konsistensi antar komponen model dalam kriteria sangat valid, (3) validitas komponen perangkat pembelajaran dalam kriteria sangat valid, (4) kepraktisan (keterlaksanaan model) pada uji coba terbatas dan luas dalam kriteria sangat baik, (5) keterampilan metakognisi siswa pada uji coba terbatas dan luas meningkat secara signifikan dengan n-gain konsisten pada kategori gain tinggi, (6) pemahaman konsep siswa pada uji coba terbatas dan luas meningkat secara signifikan dengan konsisten pada kategori n-gain tinggi, (7) dampak pengiring retensi siswa terhadap konsep-konsep yang telah dipelajari dengan kategori sangat baik dan karakter siswa pada uji coba terbatas dan luas meningkat secara signifikan dengan konsisten pada kategori n-gain tinggi, dan (8) siswa memberikan respon positif terhadap perangkat pembelajaran dan proses pembelajaran pada uji coba terbatas. Disimpukan bahwa model i-SMART yang dikembangkan adalah valid, praktis, dan efektif sehingga layak untuk meningkatkan keterampilan metakognisi dan pemahaman konsep siswa.

ABSTRACT
Syahmani. 2018. Developing of i-SMART Learning Model Using Chemistry Problem Solving in High School to Improve Students’ Metacognitive Skills and Understanding Concepts. Dissertation, Science Education Study Program, Postgraduate of Surabaya State University. Supervisors: Prof. Dr. Suyono, M. Pd. and Co-Supervisors: Dr. Z.A. Imam Supardi, M. Si.
Key words: i-SMART Learning Model, Problem Solving, Metacognitive Skills, Understanding of Chemistry

The aim of the study was to produce a valid, practical, and effective i-SMART model so that it is feasible to improve metacognition skills, character, conceptual understanding, and student retention. This research was conducted in three stages, namely the introduction phase, development, and model testing. The trial design used pre-experiment with one group pretest-posttest design. The subjects of the trial were limited to 25 students Senior High School in Banjarmasin, while the broad trial subjects consisted of 90 students from three Senior High Schools in Banjarmasin. Data collection through assessment of expert validity, tests, observations, questionnaires, interviews, and documentation. Data analysis techniques were descriptive quantitative and qualitative, n-gain, paired t test/ Wilcoxon test and Kruskal Wallis test. The results showed that: (1) the validity of the contents of the i-SMART model in terms of the need for model development and the novelty of scientific knowledge was very valid, (2) the construct validity of the i-SMART model in terms of mutual consistency between component models in very valid criteria, (3) the validity of the components of the learning device in very valid criteria, (4) practicality (implementation of the model) in limited and wide trials in very good criteria, (5) students' metacognition skills in the limited and wider trials increased significantly with n-gain consistent in the high gain category, (6) the understanding of students' concepts in the limited and wider trials increases significantly with n-gain consistent in the high gain category. (7) nurturance effect student retention of concepts that have been studied in a very good category and students' character in the limited and the wider trial increase significantly with n-gain consistent in the high gain category, and (8) students give a positive response to the learning package and process in the limited trial and the wider. The conclusion is that the i-SMART model developed is valid, practical, and effective so that it is feasible to improve students’ metacognition skills and conceptual understanding.

ABSTRAK
Syahmani. 2018. Pengembangan Model Pembelajaran Konstruktivis i-SMART Menggunakan Pemecahan Masalah Kimia SMA untuk Melatih Keterampilan Metakognisi dan Pemahaman Konsep Siswa. Disertasi, Program Studi S3 Pendidikan Sains, Pascasarjana Universitas Negeri Surabaya. Promotor: Prof. Dr. Suyono, M. Pd. dan Kopromotor: Dr. Z.A. Imam Supardi, M. Si.
Kata-kata kunci: Pembelajaran konstruktivis i-SMART, pemecahan masalah, keterampilan metakognisi, pemahaman konsep

Tujuan penelitian adalah menghasilkan model i-SMART yang valid, praktis, dan efektif sehingga layak untuk meningkatkan keterampilan metakognisi, pemahaman konsep, dan retensi siswa. Penelitian ini dilakukan dalam tiga tahap, yaitu fase pendahuluan, pengembangan, dan pengujian model. Desain uji coba menggunakan pra-eksperimen dengan one group pretest-posttest design. Subyek uji coba terbatas pada 25 siswa SMA Negeri 2 Banjarmasin, sedangkan subjek uji coba luas terdiri 90 siswa dari tiga SMA Negeri di Banjarmasin. Pengumpulan data melalui penilaian validitas pakar, tes, pengamatan, angket, wawancara, dan dokumentasi. Teknik analisis data dilakukan secara deskriptif kuantitatif dan kualitatif, n- gain, uji t berpasangan/uji Wilcoxon, dan uji Kruskal Wallis. Hasil penelitian menunjukkan bahwa: (1) validitas isi model i-SMART ditinjau dari kebutuhan akan pengembangan model dan kebaruan pengetahuan ilmiah dalam kriteria sangat valid, (2) validitas konstruk model i-SMART ditinjau dari saling konsistensi antar komponen model dalam kriteria sangat valid, (3) validitas komponen perangkat pembelajaran dalam kriteria sangat valid, (4) kepraktisan (keterlaksanaan model) pada uji coba terbatas dan luas dalam kriteria sangat baik, (5) keterampilan metakognisi siswa pada uji coba terbatas dan luas meningkat secara signifikan dengan n-gain konsisten pada kategori gain tinggi, (6) pemahaman konsep siswa pada uji coba terbatas dan luas meningkat secara signifikan dengan konsisten pada kategori n-gain tinggi, (7) dampak pengiring retensi siswa terhadap konsep-konsep yang telah dipelajari dengan kategori sangat baik dan karakter siswa pada uji coba terbatas dan luas meningkat secara signifikan dengan konsisten pada kategori n-gain tinggi, dan (8) siswa memberikan respon positif terhadap perangkat pembelajaran dan proses pembelajaran pada uji coba terbatas. Disimpukan bahwa model i-SMART yang dikembangkan adalah valid, praktis, dan efektif sehingga layak untuk meningkatkan keterampilan metakognisi dan pemahaman konsep siswa.

ABSTRACT
Syahmani. 2018. Developing of i-SMART Learning Model Using Chemistry Problem Solving in High School to Improve Students’ Metacognitive Skills and Understanding Concepts. Dissertation, Science Education Study Program, Postgraduate of Surabaya State University. Supervisors: Prof. Dr. Suyono, M. Pd. and Co-Supervisors: Dr. Z.A. Imam Supardi, M. Si.
Key words: i-SMART Learning Model, Problem Solving, Metacognitive Skills, Understanding of Chemistry

The aim of the study was to produce a valid, practical, and effective i-SMART model so that it is feasible to improve metacognition skills, character, conceptual understanding, and student retention. This research was conducted in three stages, namely the introduction phase, development, and model testing. The trial design used pre-experiment with one group pretest-posttest design. The subjects of the trial were limited to 25 students Senior High School in Banjarmasin, while the broad trial subjects consisted of 90 students from three Senior High Schools in Banjarmasin. Data collection through assessment of expert validity, tests, observations, questionnaires, interviews, and documentation. Data analysis techniques were descriptive quantitative and qualitative, n-gain, paired t test/ Wilcoxon test and Kruskal Wallis test. The results showed that: (1) the validity of the contents of the i-SMART model in terms of the need for model development and the novelty of scientific knowledge was very valid, (2) the construct validity of the i-SMART model in terms of mutual consistency between component models in very valid criteria, (3) the validity of the components of the learning device in very valid criteria, (4) practicality (implementation of the model) in limited and wide trials in very good criteria, (5) students' metacognition skills in the limited and wider trials increased significantly with n-gain consistent in the high gain category, (6) the understanding of students' concepts in the limited and wider trials increases significantly with n-gain consistent in the high gain category. (7) nurturance effect student retention of concepts that have been studied in a very good category and students' character in the limited and the wider trial increase significantly with n-gain consistent in the high gain category, and (8) students give a positive response to the learning package and process in the limited trial and the wider. The conclusion is that the i-SMART model developed is valid, practical, and effective so that it is feasible to improve students’ metacognition skills and conceptual understanding.