The Effect of Multiple Representation Based Instruction on Mathematical Achievement: A Meta-Analysis

Abstract

The aim of this study is to determine the overall effect of multiple representation based instructionon mathematical achievement. Meta-analysis method was used for this aim. The sample of the studyconsisted of 33 experimental studies within 10 publications which were selected according to certaincriteria. In data analysis, mean effect size of the selected studies was examined. The effect sizedistribution of the selected studies demonstrated a heterogeneous structure which led to apreference for random effects model (Q=66.320; p=.000).The data of the current study were analyzedusing Comprehensive Meta-Analysis (CMA), which is a specialized statistical software. The result ofthe Z-test revealed a statistically significant effect size (Z=7.015; p<.05). In order to find evidence forreliability, both a graphic approach (funnel plot) and Orwin's fail-safe N method were used to assessthe potential publication bias. Findings of these tests suggested no bias in the data. As a result, thecurrent investigation revealed that there was an overall medium and positive relationship betweenmultiple representation based instruction and mathematical achievement.