Guttman's mapping sentence technique is examined as a mechanism for defining domains of cognitive behavior and for generating test items to measure achievement in those domains. The utility of the mechanism as compared to alternatives is discussed. (Author/JKS)

The G index is a measure of similarity between pairs of dichotomized items. The G index is generalized here to the case where items are trichotomized. (JKS)

The simultaneous assessment of differential item functioning (DIF) for a collection of test items through an index that measures the variance of DIF on a test as an indicator of the degree to which different items show DIF in different directions is proposed and evaluated through simulations. (SLD)

Analysis of test item banks in 10 auditing textbooks found that 75% of questions violated one or more guidelines for multiple-choice items. In comparison, 70% of a certified public accounting exam bank had no violations. (SK)

Describes an extension to the Rasch model for fundamental measurement in which there is parameterization not only for examinee ability and item difficulty but also for judge severity. Discusses variants of this model and judging plans, and explains its use in an empirical testing situation. (SLD)

Presents empirically based item selection guidelines for moving the cut score on equated tests consisting of "n" dichotomous items calibrated assuming the Rasch model. Derivations of lemmas that underlie the guidelines are provided as well as a simulated example. (SLD)

The mixture index of fit (T. Rudas et al, 1994) is used to estimate the fraction of a population for which differential item functioning (DIF) occurs, and this approach is compared to the Mantel Haenszel test of DIF. The proposed noniterative procedure provides information about data portions contributing to DIF. (SLD)

Presents two mathematical programming approaches for the assembly of ability tests from item pools calibrated under a multidimensional item response theory model. Item selection is based on the Fisher information matrix. Illustrates the method through empirical examples for a two-dimensional mathematics item pool. (SLD)

Examined the effectiveness of the Sympson-Hetter technique and rotated content balancing relative to no exposure control and no content rotation conditions in a computerized adaptive testing system based on the partial credit model. Simulation results show the Sympson-Hetter technique can be used with minimal impact on measurement precision,…

Studied the impact of systematic item feature variation on item statistical characteristics and the degree to which such information could be used as collateral information to supplement examinee performance data and reduce pretest sample size by generating 2 families of 48 word problem variants for the Graduate Record Examinations. Results with…

Used a simulation to investigate the applicability to computerized adaptive test data of a differential item functioning (DIF) analysis method. Results show the performance of this empirical Bayes enhancement of the Mantel Haenszel DIF analysis method to be quite promising. (SLD)

Following a brief discussion of test construction by linear programing, the results of a study by F. B. Baker and others (1988) with respect to a uniform target is replicated. It is demonstrated that the result depends on characteristics of the item pool. (SLD)

An estimate and upper-bound estimate for the reliability of a test composed of binary items is derived from the multidimensional latent trait theory of R. D. Bock and M. Aitken (1981). The practical uses of such estimates are discussed. (SLD)

Linear programing was used to select items from item pools (N=1,500) based on one-, two-, and three-parameter models so that a target test information function (TTIF) was reached. Focus was on the distributional characteristics of selected items. The linear-programing approach focuses on the worst feature of the TTIF. (TJH)

Investigated the overlap between the Wechsler Intelligence Scale for Children - Revised (WISC-R) and the Halstead-Reitan Neuropsychological Battery (HRNB) in light of their use in diagnosing children's learning problems using scores for children (N=1,181) on the WISC-R and the HRNB. Results showed primary overlap between measures was attributed to…