This report proposes an empirical Bayes approach to the problem of equating scores on test forms taken by very small numbers of test takers. The equated score is estimated separately at each score point, making it unnecessary to model either the score distribution or the equating transformation. Prior information comes from equatings of other…

Although partial correlation is a correlation of residuals, the correlation of the true-score components of these residuals is not equivalent to the partial correlation of the true scores themselves. The source of this discrepancy is explained and its implications are briefly discussed. (Author)

Article commented on a study by Harris, who presented formulas for the variance of errors of estimation (of a true score from an observed score) and the variance of errors of prediction (of an observed score from an observed score on a parallel test). (Author/RK)

The procedure of estimating true scores by means of a transformation of the obtained score based on the reliability coefficient is compared with the use of the obtained score without transformation. Using the mean squared error as a criterion, the transformed score is a better estimate for most examinees but poorer for those whose true scores lie…

A distinction is made between reliability of measurement and reliability of classification; the "criterion-referenced reliability coefficient" describes the former. Application of this coefficient to the probability distribution of possible scores for a single student yields a meaningful way to describe the reliability of a single score. (Author)

The traditional reliability coefficient and standard error of measurement are not adequate measures of reliability for tests used to make pass/fail decisions. Answering the important reliability questions requires estimation of the joint distribution of true and observed scores. Lord's "Method 20" estimates this distribution without the…

Besides the ubiquitous Pearson product-moment r, there are a number of other measures of relationship that are attenuated by errors of measurement and for which the relationship between true measures can be estimated. Among these are the correlation ratio (eta squared), Kelley's unbiased correlation ratio (epsilon squared), Hays' omega squared,…

Much previously published material for estimating the reliability of classification has been based on the assumption that a test consists of a known number of equally weighted items. The test score is the number of those items answered correctly. These methods cannot be used with classifications based on weighted composite scores, especially if…

This article is a reply to a previous paper (see TM 500 488) interpreting Livingston's original article (see TM 500 487). (CK)

The assumptions of the classical test-theory model are used to develop a theory of reliability for criterion-referenced measures which parallels that for norm-referenced measures. It is shown that the Spearman-Brown formula holds for criterion-referenced measures and that the criterion-referenced reliability coefficient can be used to correct…

Procedures are described for studying the reliability of decisions based on specific passing scores with tests made up of discrete items and designed to measure continuous rather than categorical traits. These procedures are based on the estimation of the joint distribution of true scores and observed scores. (CTM)

A reliability coefficient for criterion-referenced tests is developed from the assumptions of classical test theory. The coefficient is based on deviations of scores from the criterion score, rather than from the mean. (Author/CK)