[English]: This study examined the effect of Problem-based Learning (PBL) on students' creative mathematical thinking in Indonesia during the last eight years using meta-analysis. Data were obtained from primary studies published in national journals, conference proceedings, and master thesis from 2012 to 2020. This study analyzed 19 effect sizes from 19 primary studies that fulfill the inclusion criteria. The effect size index used the Hedges' g equation and statistical analysis was assisted by comprehensive meta-analysis (CMA) software. The effect size variation test was conducted by analyzing the study's five characteristics, namely the year of study, level of education, sample size, length of treatment, and publication sources. Based on the analysis results, the overall effect size was 0.821, with a standard error of 0.130 at the 95% confidence interval (CI) referring to the random-effects model. The effect size is classified as moderate, meaning that PBL has a moderately positive effect on students' creative mathematical  thinking. The results of the analysis of the characteristics of the study show that the application of PBL to improve students' mathematical creative thinking relates to the research year (Q = 55.683; p <0.05), the treatment duration (Q = 5.644; p <0.05), and publication sources (Q = 13.812; p <0.05). It was found that the PBL was more effective in conditions of treatment duration of fewer than four meetings. There is a tendency that PBL has a significant impact when it is first implemented. The findings' implications are discussed to provide important information as a basic idea for further PBL research and implementation.

Keywords: Problem-based learning, Creative thinking, Meta-analysis

[Bahasa]: Penelitian ini menguji pengaruh problem-based learning (PBL) terhadap berpikir kreatif matematis siswa di Indonesia selama delapan tahun terakhir melalui meta-analisis. Data diperoleh dari studi primer yang telah dipublikasikan di jurnal nasional, prosiding konferensi, dan tesis master tahun 2012-2020. Penelitian ini menganalisis 19 ukuran efek dari 19 studi primer yang memenuhi kriteria penyertaan. Indeks ukuran efek menggunakan persamaan Hedges' g dan analisis statistik dibantu oleh perangkat lunak comprehensive meta-analysis (CMA). Pengujian variasi ukuran efek dilakukan dengan menganalisis lima karakteristik penelitian yaitu tahun studi, tingkat pendidikan, ukuran sampel, lama perawatan, dan sumber publikasi. Berdasarkan hasil analisis, ukuran efek keseluruhan adalah 0,821, dengan kesalahan standar 0,130 pada interval kepercayaan 95% (CI) menurut model efek acak. Ukuran efek tergolong sedang, artinya penggunaan PBL berpengaruh positif sedang terhadap berpikir kreatif matematis siswa. Hasil analisis karakteristik penelitian menunjukkan bahwa penerapan PBL untuk meningkatkan berpikir kreatif matematis siswa berkaitan dengan tahun penelitian (Q = 55,683; p <0,05), durasi perlakuan (Q = 5,644; p <0,05) , dan sumber publikasi (Q = 13.812; p <0.05). Ditemukan bahwa PBL lebih efektif pada kondisi durasi perlakuan kurang dari 4 pertemuan. Terdapat kecenderungan bahwa PBL berdampak besar saat pertama kali diterapkan. Implikasi hasil penelitian dibahas untuk memberikan informasi penting sebagai ide dasar untuk penelitian dan implementasi PBL selanjutnya.

Kata kunci: Problem-based learning, Berpikir kreatif, Meta-analysis