ERIC Number: ED445036
Record Type: Non-Journal
Publication Date: 2000-Apr
Reference Count: N/A
Applications of the Analytically Derived Asymptotic Standard Errors of IRT Item Parameter Estimates.
Li, Yuan H.; Lissitz, Robert W.
The analytically derived expected asymptotic standard errors (SEs) of maximum likelihood (ML) item estimates can be predicted by a mathematical function without examinees' responses to test items. The empirically determined SEs of marginal maximum likelihood estimation/Bayesian item estimates can be obtained when the same set of items is repeatedly estimated from test data. Understanding the consistency of SEs yielded from both approaches is of primary concern for the application of the analytic SEs. In most cases in the simulation conducted, SEs yielded from both approaches were similar, especially for the generalized partial credit model (E. Muraki, 1992). This finding encourages test practitioners and researchers to apply the asymptotic SEs of item estimates to the following applications: (1) practical testing situations; (2) item-linking studies; and (3) predicting the SEs of equating scores for the item response theory (IRT) true-score procedures (Lord, 1982) without examinees' responses to test items. Three dimensional graphical representation of the analytic SEs of item estimates has also been provided for better understanding of several widely used IRT models. (Contains 32 references.) (Author/SLD)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: N/A
Authoring Institution: N/A