Effectiveness of Problem-Based Learning Model on Students^ Mathematical Representation Skills: A Meta-Analysis Novia Permata Barti, Dadang Juandi, Dian Usdiyana
Universitas Pendidikan Indonesia
Abstract
The present study aims to investigate the effectiveness of the implementation of Problem-Based Learning Model on the students^ mathematical representation skills comprehensively. The meta-analysis method was employed to analyze the effect size of primary studies. The empirical data were obtained through electronic database searches of primary studies published in national journals between 2011 and 2020. Nine effect sizes were analyzed from nine primary studies that met the inclusion criteria. The data analysis was conducted using the Comprehensive Meta-Analysis (CMA) software, by selecting the Hedges^ g formula to determine the effect size based on the random effects model. The publication bias and sensitivity analysis resulted in that all studies included in the analysis were not susceptible, meaning resistant to the publication bias. The effect size test was conducted by analyzing three characteristics, i.e., grade level, sample size, and treatment duration. The findings show that, overall, the effect of the implementation of Problem Based Learning model on the students^ mathematical representation skills is better than that of conventional learning, with the level of influence found in the medium category. Meanwhile, based on the characteristics of the observed studies, there is no difference in the effect between groups in terms of grade level, sample size, and treatment duration. Therefore, this study contributes as a literature for consideration in implementing the Problem-Based Learning in determining learners^ mathematical representation skills in the future.
