The Effect of Scratch-Based Instructional Media on Grade 5 Elementary Students’ Coding Skills: A Quasi-Experimental Study

The integration of coding education into the elementary curriculum is crucial for nurturing computational thinking from an early age. However, many students still struggle to understand abstract programming concepts through conventional teaching methods. This study aimed to investigate the impact of Scratch-based instructional media on the coding skills of Grade 5 students, specifically their ability to design, debug, and execute block-based programs. Using a quasi-experimental non-equivalent control group design, the research was conducted at SDN Bangsalsari 02, Jember, Indonesia, involving 75 students (38 in the experimental class using Scratch, and 37 in the control class using conventional instruction). Data were gathered using validated pretest–posttest instruments that assessed algorithmic thinking and problem-solving in coding. Statistical analysis employed paired-sample and independent-sample t-tests with effect size calculations. The results revealed that the experimental group's mean score increased significantly from 19.20 to 30.50 (Cohen's d = 1.85, a significant effect), while the control group improved from 21.60 to 28.60 (Cohen's d = 0.92, a moderate effect). Posttest comparisons confirmed that the experimental group outperformed the control group (t(73) = 2.13, p = .037, d = 0.49, moderate effect size). These findings demonstrate that Scratch-based learning significantly enhances elementary students' coding competence by providing a visual, interactive, and engaging learning environment. The study highlights Scratch's potential as an effective tool for developing computational thinking within Indonesia's elementary education context and recommends further research on its long-term pedagogical impacts.