Abstract
Understanding how instructional design shapes differentiated cognitive outcomes remains a central question in educational psychology, particularly in culturally grounded learning contexts where heritage knowledge must be retained, interpreted, and applied. Existing research on inquiry-based learning (IBL) in heritage education has relied largely on qualitative or non-comparative designs, and heritage cognition has typically been assessed through aggregate measures that may obscure important variation across cognitive levels. This study addresses these gaps by examining whether IBL produces differentiated cognitive effects across hierarchical levels in tea cultural heritage (TCH) education. A quasi-experimental design was employed with 66 seventh-grade pupils in the Wuyi Mountain area, Fujian Province, China, a UNESCO World Heritage site with deep historical ties to tea production. Two intact classes were assigned to different conditions: the experimental class received a purpose-designed IBL module structured around six inquiry stages—questioning, data collection, data analysis, result organization, sharing, and reflection—whereas the control class received conventional teacher-led instruction. Cognitive outcomes were assessed using a seven-level hierarchical framework adapted from Bloom’s taxonomy, spanning knowledge, comprehension, passive application, active application, analysis, synthesis, and evaluation. This framework was operationalized through a 100-point cognitive test combining multiple-response items for knowledge, comprehension, analysis, and evaluation with rubric-scored open-ended tasks for passive application, active application, and synthesis. Between-group and within-group differences were examined using non-parametric analyses. Results revealed a statistically significant overall cognitive advantage for the IBL condition, with a large effect size (
r
= 0.769). These gains were not uniformly distributed across the cognitive hierarchy. The strongest between-group contrasts emerged at application and synthesis (
r
= 0.816–0.831), where both groups had near-zero baseline scores. These immediate findings should be interpreted cautiously because the conditions differed in facilitation, context, and response format, and because the study involved two intact classes. The study demonstrates the value of hierarchical assessment, and provides quasi-experimental evidence concerning a contextually grounded IBL package in heritage education.