Integrating scratch-based didactic design into probability learning: An innovative pathway toward sustainable development goals in middle school mathematics

Department of Mathematics Education, Universitas Negeri Yogyakarta, Yogyakarta, Indonesia

^* Corresponding author: sitikholifah@uny.ac.id

Published online: 22 December 2025

Abstract

The implementation of Sustainable Development Goal 4 requires innovative pedagogical approaches in mathematics education, particularly in developing countries where traditional methods have proven insufficient in fostering deep conceptual understanding of probability. Despite its importance, probability remains one of the most challenging topics for middle school students due to abstract concepts and limited engagement in conventional instruction. The research contribution is providing empirical evidence on the effectiveness of Scratch-based digital didactic design as an accessible and inclusive learning approach that enhances probability comprehension while developing computational thinking skills aligned with SDG 4 quality education objectives. This study employed an experimental pre-test and post-test control group design involving 62 eighth-grade students from a secondary school in Yogyakarta, Indonesia, divided into experimental (n = 31) and control groups (n = 31). The experimental group received instruction using Scratch programming applications integrated with probability concepts, while the control group followed conventional teaching methods. Data were collected through validated probability comprehension tests administered before and after the eight-week intervention. Results demonstrated significant superiority of the Scratch-based approach, with the experimental group achieving an average normalized gain of 0.71 (high category) compared to 0.42 (moderate category) in the control group. Statistical analysis revealed substantial improvements in the experimental group (t = −8.92, p < 0.001) versus moderate gains in the control group (t = −3.45, p < 0.001). The independent t-test confirmed significant differences between groups (t = −4.797, p < 0.001) with a large effect size (Cohen's d = 1.22), indicating meaningful practical impact. This study concludes that integrating visual programming environments into mathematics education significantly enhances conceptual understanding and supports SDG 4 through innovative digital learning approaches.