Performance-Based Assessment of Scientific Creativity among Primary School Students: Framework Development and Practical Verification

HRMARS - Purpose: Scientific creativity is a core component of scientific literacy, yet traditional standardized assessments show poor adaptability for primary school students and fail to capture process-based creative performance. Grounded in progressive education principles, this study aims to develop a performance-based assessment system for primary school students’ scientific creativity and verify its effectiveness, so as to fill the gap of integrated assessment framework in the field of primary science evaluation. Methodology: A mixed-methods design was adopted. A three-tier assessment framework was constructed by integrating scientific creativity theory and Williams’ creativity model (Williams, 1980). Two instruments—a 40-item Scientific Creativity Scale and a 20-item Influencing Factors Questionnaire—were developed and tested with 180 Grade 4 students from Sichuan, China. Implementation strategies for classroom practice were refined through action research. Findings: First, the assessment framework comprises 3 first-level dimensions, 13 second-level dimensions and 27 behavioral indicators. Second, the scale achieves good reliability (Cronbach’s ? = 0.897) and validity. Third, a three-stage implementation strategy and a five-step instructional design path were proposed for science and maker courses. Fourth, students’ overall scientific creativity is above average, with higher scores in creative tendency and lower scores in divergent thinking; significant gender differences exist in curiosity and science learning interest. Implications: This study enriches the theoretical system of performance-based assessment in science education, and provides operable practical tools for progressive science education, supporting the shift from result-oriented to process-oriented evaluation in primary science classrooms.