Identification of Computational Thinking Patterns in Students’ Mathematical Problem Solving: Empirical Evidence from Algebra Learning Activities

Abstract

Computational thinking (CT) is increasingly recognized as a core competency in STEM education, yet little empirical evidence exists on how CT patterns naturally emerge in algebraic reasoning outside programming contexts. This study investigates how secondary students demonstrate CT components during algebra problem-solving. A mixed-methods design was applied with 120 students completing algebra tasks. Data were collected through assessments, interviews, and classroom observations. The elements of CT, namely pattern recognition, decomposition, abstraction, and algorithmic reasoning, are systematically coded and analyzed using qualitative and quantitative approaches. Findings revealed that pattern recognition (80%) and decomposition (70%) were widely observed across achievement levels. In contrast, abstraction (45%) and algorithmic reasoning (35%) appeared more frequently among high-achieving students. Statistical analysis confirmed significant differences in CT sophistication across achievement groups, highlighting progression from basic recognition to advanced reasoning. The novelty of this study lies in its empirical demonstration of CT within algebraic problem-solving, independent of programming environments. Unlike prior research emphasizing coding, it shows how CT components naturally emerge in mathematics tasks. By analyzing achievement-level differences, it provides fresh evidence of developmental trajectories in CT. These results position algebra as a strategic entry point for CT development, bridging mathematical reasoning and computational approaches. They suggest that intentional pedagogical design can foster advanced CT skills in conventional classrooms, offering practical guidance for curriculum innovation in mathematics education.