Integration of Item Hierarchy and Concept Tree based on Clustering Approach with Application in Statistics Learning

Item hierarchy and concept tree provide references for cognition diagnosis and remedial instruction. Therefore, integration of data analysis on item hierarchy and concept tree should be important. The purpose of this study is to provide an integrated methodology of item hierarchy and concept tree analysis. Besides, fuzzy clustering is adopted to classify sample so that homogeneity appear in the same cluster and adaptive instruction will be more feasible. Polytomous item relational structure (PIRS) is the foundation of item hierarchy analysis. Interpretive structural modeling (ISM) combined with calculation of ordering coefficient is to construct concept tree. Source data sets of PIRS and ISM are based on response data matrix and item-attribute matrix respectively. In this study, the empirical test data is the statistics assessment of university students. The results show that the integration of PIRS and ISM based on fuzzy clustering are useful for cognition diagnosis and adaptive instruction. Finally, further suggestions and recommendations based on findings are discussed. Key-Words: cognition diagnosis, concept tree, fuzzy clustering, ISM, item hierarchy, PIRS