Creative problem-solving assessment in fluid mechanics for senior high school students: Instrument validation and reliability analysis using the rasch model

Background: Creative Problem-Solving (CPS) is a critical skill in physics education, enabling students to address real-world challenges through innovative and practical approaches. Despite its importance, there is a limited number of reliable and valid instruments specifically designed to assess CPS skills within the context of physics education. Aim: This study aims to develop and prove the validation of a CPS assessment instrument using the Rasch model to ensure its reliability and accuracy in measuring CPS skills in physics education. Method: The research followed the ADDIE development model, which includes analysis, design, development, implementation, and evaluation phases. The instrument, consisting of 20 essay items aligned with CPS indicators, was validated by three experts in physics education. It was then tested on 35 Grade XI students in Lampung Province. Data analysis was performed using the Rasch model through Winsteps software, focusing on fit analysis, reliability, item difficulty distribution, and dimensionality assessment. Results: The instrument demonstrated strong internal consistency, confirming its reliability for assessing CPS skills in physics education. The construct validity was supported by fit analysis, indicating that most items functioned appropriately. However, some items exhibited potential bias and required revision. The dimensionality analysis confirmed that the instrument effectively measured a single underlying construct, ensuring its psychometric robustness. These findings suggest that the developed CPS assessment instrument is a reliable and valid tool for evaluating students’ creative problem-solving abilities in high school physics, particularly in fluid mechanics. Conclusion: This study successfully developed and validated a CPS assessment instrument for high school physics education, demonstrating strong reliability and construct validity. The instrument effectively measures CPS skills, though some items require refinement to ensure fairness and accuracy. Future research should focus on further improving item quality and testing the instrument in diverse educational settings to enhance its applicability and generalizability.