ABSTRACT

This study examines the psychometric validity and reliability of an achievement test designed for a fusion-based suona technique curriculum in Chinese universities. Grounded in constructivist theory, Vygotsky’s zone of proximal development, and multiple intelligences theory, the assessment instrument was developed to measure diverse competencies ranging from theoretical knowledge to performance application. Utilizing a post-test-only design to collect validation data from an experimental cohort, the study employed a dual-framework psychometric analysis. Data were obtained from a 40-item multiple-choice test and performance-based tasks. Psychometric evaluation revealed a generally balanced distribution of item difficulty; however, Classical Test Theory (CTT) analysis indicated that while 25% of items demonstrated excellent discrimination, 50% required revision to improve distractor efficiency. Complementary Item Response Theory (IRT) analysis, using the Rasch rating scale model, demonstrated satisfactory internal consistency (Person Reliability = .776) and acceptable model-data fit as well as local independence (MADaQ3 p = .182). Wright Map analysis confirmed appropriate targeting for average-ability students but highlighted a need for additional items to adequately measure high-ability levels. These findings provide empirical evidence for the instrument’s utility and offer a framework for refining assessment tools in multicultural and competency-based music pedagogy.

Keyword :