Effectiveness of Constructivism Based Learning Models Against Students Mathematical Creative Thinking Abilities in Indonesia; A Meta-Analysis Study

Abstract

The purpose of this meta-analysis review is to examine the effectiveness of the application of constructivism-based learning models to the mathematical creative thinking abilities of secondary school students in Indonesia. This meta-analysis analyzes 33 effect sizes from 33 primary studies that met the inclusion criteria and published them in ERIC documents, journal articles, and proceedings from 2010 to 2019. Data were analyzed using CMA (Compherensive Meta-Analysis) software designed to determine publication bias and effect size. As a result of the analysis, the combined effect size was 0.884 based on the random-effects model. These results can be interpreted that the average student who was ranked 13th in the experimental class, the same as students who were ranked 6th in the control class. The results of the analysis of the mediator variables found that all variables are related to the effect size. The strongest relationship was found for the research class variables and sample size. There is evidence that constructivism-based learning models are more effective if applied in conditions of a sample size of 1-30 students. These facts are considered by mathematics education practitioners in applying constructivism-based learning models in the future.