| Abstrakt: |
When attributes are hierarchically structured, modifying the Q-matrix or prior distribution in the estimation process yields more accurate and precise item and person parameter estimates. Modification of the prior distribution and the Q-matrix depend on the assumed attribute hierarchy, as such, identifying the correct hierarchical structure among a set of measured is of the essence. To address the subjectivity in the conventional methods for attribute structure identification (i.e., expert opinions via content analysis and verbal data analyses such as interviews and think-aloud protocols); this study proposes a likelihood-ratio-test based exhaustive empirical search method for identifying hierarchical structures. It further suggests employment of likelihood-ratio-test based model selection approach for choosing the most accurate hierarchical structure among proposed candidates. Results of this study show that the likelihood-ratio-test based exhaustive search produces a reachability matrix that specifies all the true prerequisite relationships among the attributes. Thus, the method is promising and may be used for exploratory purposes for identification of hierarchical attribute structure. [ABSTRACT FROM AUTHOR] |
