The Use of Eliciting Activities Model to Improve Mathematical Representation Ability and Self-Efficacy in Elementary Schools

This study aims to determine improvements in mathematical representation ability and self-efficacy in ratio material among fifth-grade elementary school students using Model-Eliciting Activities (MEAs). This research employs a mixed-method approach with a quasi-experimental nonequivalent control group design. The experimental group received treatment using Model-Eliciting Activities (MEAs), while the control group was taught using conventional learning methods. Data collection techniques in this study included tests, questionnaires, documentation, observation, and interviews. Data analysis consisted of descriptive statistics, prerequisite tests, and final hypothesis testing (T-test, Z-test, mean difference test, and gain test). The results of this study indicate that Model-Eliciting Activities (MEAs) have a significant effect on the mathematical representation ability of fifth-grade students in the Sultan Agung cluster in the ratio material. This is demonstrated by the analysis results using the paired sample t-test, which produced a t-value (t_calculated) greater than the t-table value (13.047 > 2.145). Model-Eliciting Activities (MEAs) are effective in improving students’ mathematical representation ability, as shown by the independent sample t-test results, where t_calculated was greater than the t-table value (2.52 > 2.048). There is a significant difference in self-efficacy between students who received MEA-based learning and those who received conventional learning. The analysis using the independent samples t-test showed that the t-calculated values for self-efficacy were 2.935, 2.522, 2.699, and 2.871 for the magnitude, strength, and generality dimensions, respectively. Given that the t-table value was 2.048, it can be concluded that the t-calculated values for the self-efficacy questionnaire results in the experimental and control classes, across the dimensions of magnitude, strength, and generality, were greater than the t-table value.