Coding-based instruction for developing computational thinking in primary education: a systematic review

Abstract

The integration of computational thinking (CT) into elementary education is increasingly viewed as essential for developing students' problem-solving and digital literacy skills. However, effective implementation of CT-based instruction through coding remains varied in practice. This study aims to synthesise empirical evidence on how coding supports CT development in primary school students. A systematic literature review was conducted following PRISMA guidelines, with an initial identification of 1145 records from Scopus and Google Scholar. After removing 104 duplicates, 1041 titles and abstracts were screened, and 113 full-text articles were assessed for eligibility. Based on the inclusion criteria, five empirical studies published between 2021 and 2025 were selected. Thematic analysis revealed that instructional approaches such as unplugged activities, educational robotics, and pair programming effectively enhanced core CT skills—abstraction, decomposition, and algorithmic thinking—while also promoting student engagement and collaboration. However, persistent barriers include limited teacher readiness, inadequate assessment tools, and unequal access to technology. The study concludes that early exposure to structured coding interventions positively influences CT development. To ensure sustainability and equity, future efforts must focus on teacher training, culturally responsive pedagogy, and integration of CT into the broader curriculum. This review provides a focused synthesis that can guide educational stakeholders in designing and supporting effective CT instruction at the primary level.