In behavioural sciences, local dependence and DIF are common, and purification procedures that eliminate items with these weaknesses often result in short scales with poor reliability. Graphical loglinear Rasch models (Kreiner & Christensen, in "Statistical Methods for Quality of Life Studies," ed. by M. Mesbah, F.C. Cole & M.T.…

Multidimensional computer adaptive testing (MCAT) can provide higher precision and reliability or reduce test length when compared with unidimensional CAT or with the paper-and-pencil test. This study compared five item selection procedures in the MCAT framework for both domain scores and overall scores through simulation by varying the structure…

Parameter estimation for Keats generalization of the Rasch model that takes account of guessing behavior is investigated. It is shown that no minimal sufficient statistics for the ability parameters independent of the difficulty parameters exist. (Author/JKS)

Use of the same term "split-half" for division of an n-item test into two subtests containing equal (Cronbach), and possibly unequal (Guttman), numbers of items sometimes leads to a misunderstanding about the relation between Guttman's maximum split-half bound and Cronbach's coefficient alpha. This distinction is clarified. (Author/JKS)

A modification of the method of reciprocal averages for scaling multiple choice data is proposed. The proposed method handles the data in a piecewise fashion and allows for faster convergence to a solution. (Author/JKS)

A method of estimating item characteristic functions is proposed, in which a set of test items, whose operating characteristics are known and which give a constant test information function for a wide range of ability, are used. The method is based on maximum likelihood estimation procedures. (Author/JKS)

Estimating the reliability of measures derived from separate questions on essay tests or individual judges on a rater panel is considered. Cronbach's alpha is shown to underestimate reliability in these cases. Some alternative coefficients are presented. (JKS)

Presently available test statistics for the Rasch model are shown to be insensitive to violations of the assumption of test unidimensionality. Two new statistics are presented. One is similar to available statistics, but with some improvements; the other addresses the problem of insensitivity to unidimensionality. (Author/JKS)

An important relationship is given for two generalizations of coefficient alpha: (1) Rajaratnam, Cronbach, and Gleser's generalizability formula for stratified-parallel tests, and (2) Raju's coefficient beta. (Author/CTM)

A chain of lower bound inequalities leading to the greatest lower bound to reliability is established for the internal consistency of a composite of unit-weighted scores (such as a test). Algorithms for obtaining various reliability coefficients are presented. (Author/JKS)

Measures of consistency and completeness of order relationships derived from test data such as Guttman scales are proposed. The measures are generalized to apply to incomplete data such as data from tailored testing. (Author/JKS)

On distractor-identification tests students mark as many distractors as possible on each test item. A grading scale is developed for this type testing. The score is optimal in that it yields an unbiased estimate of the student's score as if no guessing had occurred. (Author/JKS)

A procedure for determining the reliability of an examinee knowing k out of n possible multiple choice items given his or her performance on those items is presented. Also, a scoring procedure for determining which items an examinee knows is presented. (Author/JKS)

Given known item parameters for a test, unbiased estimators are derived for an examinee's ability parameter, his or her proportion correct true score for the variances of these parameters and for the parallel forms reliability of the maximum likelihood estimator of the ability parameter. (Author/JKS)