Study refers to J. A. Bowers' A note on Gaylord's Estimating test reliability from the item-test correlations,''' EJ 041 295. (CB)

The consequences of the assumption that the expected score is equal to the true score are shown and alternatives discussed. (MS)

This paper presents three different ways of computing the internal consistency coefficient alpha for a same set of data. The main objective of the paper is the illustration of a method for maximizing coefficient alpha. The maximization of alpha can be achieved with the aid of a principal component analysis. The relation between alpha max. and the…

Estimates for the variances of empirically determined scoring weights are given. It is also shown that test item writers should write distractors that discriminate on the criterion variable when this type of scoring is used. (Author)

The professional orientation difference approach was operationalized as a stress on the importance of intellectual activity, autonomy, service and influence, with de-emphasis on the monetary, prestige, security and human relations aspects of a job. As a difference score, the index was subject to special reliability problems. This analysis…

This investigation dealt with the development and evaluation of both a music dictation test (PRM78 Dictation Test) and a sightsinging test (CSS76 Criterion Sightsinging Test). It was hoped that the dictation test could eventually be developed to serve as an adequate replacement for the latter. Thirteen samples participated in this project--7…

Data from 200 college-level tests were used to compare three reliability approximations (two of Saupe and one of Cureton) to Kuder-Richardson Formula 20 (KR20). While the approximations correlated highly (about .9) with the reliability estimate, they tended to be underapproximations. The explanation lies in an apparent bias of Lord's approximation…

Two new methods for constructing a credibility interval (CI)--an interval containing a specified proportion of true validity description--are discussed, from a frequentist perspective. Tolerance intervals, unlike the current method of constructing the CI, have performance characteristics across repeated applications and may be useful in validity…

Copies of this program may be obtained from the author at the University of Southern California, School of Pharmacy, University Park, Los Angeles 90007. (CB)

The classical estimate of a binomial probability function is to estimate its mean in the usual manner and to substitute the results in the appropriate expression. Two alternative estimation procedures are described and examined. Emphasis is given to the single administration estimate of the mastery test reliability. (Author/CTM)

Generalizability theory is used to characterize and quantify expected variance in cutting scores and to compare the Nedelsky and Angoff procedures for establishing a cutting score. Results suggest that the restricted nature of the Nedelsky (inferred) probability scale may limit its applicability in certain contexts. (Author/BW)