Development and psychometric evaluation of a learning style instrument for chemistry learning

Abstract

This study aimed to develop a valid and reliable learning style instrument specifically designed for chemistry learning among senior high school students. Understanding students’ learning styles is essential in chemistry education due to the abstract nature of chemical concepts and the need for effective instructional strategies that accommodate individual learning preferences. This research employed a Research and Development (R&D) approach with a descriptive quantitative design, adapting the ten-step non-test instrument development procedure proposed by Mardapi. The instrument was developed through the adaptation of the Solomon and Felder learning style instrument and validated by two chemistry education experts. Field testing was conducted with 63 eleventh-grade students at SMA Negeri 9 Bengkulu. Data were analyzed using Product Moment correlation for validity testing and Cronbach’s Alpha for reliability testing. The results showed that out of 11 developed items, 10 items were valid with correlation coefficients exceeding the critical value (r-table = 0.248), while one item was found invalid and subsequently removed. Reliability analysis yielded a Cronbach’s Alpha coefficient of 0.62, indicating a moderate level of reliability. These findings demonstrate that the developed instrument possesses acceptable validity and reliability for identifying students’ learning styles in chemistry learning. Therefore, the instrument can be utilized as a practical tool to support teachers in designing more effective and student-centered chemistry learning strategies. Further refinement and broader-scale testing are recommended to improve the instrument’s reliability and applicability.