The first part of this paper describes the use of the robust z[subscript R] statistic to link test forms using the Rasch (or one-parameter logistic) model. The procedure is then extended to the two-parameter and three-parameter logistic and two-parameter partial credit (2PPC) models. A real set of data was used to illustrate the extension. The…

There are at least two procedures to assess item difficulty stability in the Rasch model: robust z procedure and "0.3 Logit Difference" procedure. The robust z procedure is a variation of the z statistic that reduces dependency on outliers. The "0.3 Logit Difference" procedure is based on experiences in Rasch linking for tests…

By analyzing the Fisher information allotted to the correct response of a Rasch binary item, Huynh (1994) established the response probability criterion 0.67 (RP67) for standard settings based on bookmarks and item mapping. The purpose of this note is to help clarify the conceptual and psychometric framework of the RP criterion.

Two approaches, the minimax approach and the Rasch procedure, are described for the simultaneous determination of passing scores for subtests when the passing score for the total test is known. (Author/LMO)

This study examined comparability of student scores obtained from computerized and paper-and-pencil formats of the large-scale statewide end-of-course (EOC) examinations in the two subject areas of Algebra and Biology. Evidence in support of comparability of computerized and paper-based tests was sought by examining scale scores, item parameter…

Using the Rasch model for ordered categories, this paper provides a method for qualitative interpretations of data from an affective measure such as an attitude scale. This maximum information approach provides explicit qualitative information about the meaning of a score on the latent trait being assessed. (SLD)

Each Rasch (Masters) partial credit item can be expressed as the sum of independent Rasch binary items and Rasch (or Masters) indecomposable trinary items. If each trinary item is then expressed as the sum of two binary items, the binary items are positively dependent and cannot be both Rasch types. (SLD)

The use of Cohen's kappa index as a measure of the reliability of multiple classifications is developed. Special cases of the index as well as the effects of test length on the index are also explored. (JKS)

Given a Masters partial credit item with n known step difficulties, conditions are stated for the existence of a set of (locally) independent Rasch binary items such that their raw score and the partial credit raw score have identical probability density functions. (Author/SLD)

Under the assumptions of classical measurement theory and the condition of normality, a formula is derived for the reliability of composite scores. The formula represents an extension of the Spearman-Brown formula to the case of truncated data. (Author/JAZ)

Procedures based on latent trait models and the Rasch model are described for computation and asymptotic statistical inference for two decision consistency indexes often used in mastery or competence testing. Simulations were conducted for typical mastery testing situations to illustrate the procedures. (SLD)

In the context of pass/fail decisions, using the Bock multi-nominal latent trait model for moderate-length tests does not produce decisions that differ substantially from those based on the raw scores. The Bock decisions appear to relate less strongly to outside criteria than those based on the raw scores. (Author/JAZ)