<![CDATA[Computational Thinking (CT) has become a crucial competence in 21st-century vocational education, particularly in programming-related fields. However, programming instruction in vocational high schools often emphasizes syntactic correctness rather than higher-order problem-solving processes. This study aims to analyze the effectiveness of integrating the HUTS Online Judge (HUSTOJ) system with a Problem-Based Learning (PBL) model in improving students’ CT skills and to examine students’ learning behavior patterns through system log data. The study employed a quantitative pre-experimental one-group pretest-posttest design involving 38 tenth-grade vocational students majoring in Software Engineering. CT skills were measured using Problem-Based programming tests covering decomposition, pattern recognition, abstraction, and algorithmic thinking, while learning behaviors were analyzed using HUSTOJ activity logs. The results indicate a significant improvement in CT skills, with a large effect size and a moderate-to-high normalized gain. Log analysis revealed distinct learning behavior profiles, highlighting the role of iterative refinement and persistence in CT development. These findings suggest that integrating HUSTOJ and PBL provides effective scaffolding for learning and process-oriented assessment, making it a promising instructional model for vocational programming education.]]>
Copyrights © 2026
