A Study on the Design and Practice of Pair-Programming Instruction for Junior Secondary Students Aimed at Developing Computational Thinking

With the promulgation and implementation of China’s Education Modernisation 2035 and the Compulsory Education Information Technology Curriculum Standards (2022 Edition), computational thinking has become an essential core competency for Chinese students in the digital age. However, information technology teaching in junior secondary schools across the northwest region faces numerous practical challenges, including inadequate cultivation of computational thinking, inefficient collaborative learning, and superficial learning. This study aims to investigate the impact of Pair-Programming (P-P) for secondary school pupils on the development of computational thinking, self-efficacy, and collaborative abilities, thereby providing new strategies and theoretical support for information technology teaching. Based on social constructivism, the zone of proximal development, and cognitive load theory, this study employed a quasi-experimental design. Three eighth-grade classes from municipal junior high schools in Northwest China were randomly assigned to independent, paired, and group collaborative programming groups. A 12-week instructional intervention was implemented. Utilizing a mixed-methods research paradigm combined with the Computational Thinking Scale (CTS) and other methodologies, the study systematically evaluated the paired programming strategy and its multidimensional impact on student development. Findings indicate: 1) P-P has demonstrated a positive impact on three core dimensions of junior high students’ computational thinking development, programming self-efficacy, and collaborative skills in terms of overall effectiveness. 2) Different pairing strategies exert markedly distinct effects on computational thinking and collaborative skills. Pairing arrangements should holistically consider three dimensions-programming proficiency, learning styles, and social characteristics—with proximal gradient pairing yielding optimal outcomes.