Effects of Blended Learning on Students Mathematical Problem-Solving Skills: A Meta-Analysis
Muh. Fachri B. Paloloang (a*), Dadang Juandi (b), Jarnawi Afgani Dahlan (c)

Mathematics Education, Graduate School, Universitas Pendidikan Indonesia, Bandung

Abstract

Many studies on students mathematical problem-solving skills (MPSS) have been influenced by using the blended learning (BL) model in Indonesia to date, and several reports have shown inconsistent results. Meanwhile, education policymakers need accurate and precise descriptions and information about it. To summarize, estimate, and evaluate the overall effect of using the BL model on students MPSS, a meta-analysis was used. Data obtained from primary studies that have been published in national journals during 2014-2021. The effect size index uses the Hedges g formula, and statistical analysis is assisted by comprehensive meta-analysis (CMA) software. Testing effect size variations were carried out by analyzing the five characteristics of the study, namely the year of study, level of education, sample size, sampling technique, and the research method used. Based on the analysis, the overall effect size (g) was 1.075 according to the random-effects model. The effect size is large, meaning that the use of BL has a high positive effect on students MPSS. In addition, the characteristics of educational attainment significantly lead to heterogeneous effect size data. Thus, these results recommend that Indonesian mathematics lecturers and teachers choose BL as an alternative solution to improve students MPSS.

Keywords: Blended Learning- Mathematical Problem-Solving Skills- Meta-Analysis

Topic: Mathematics Education