There are difficulties in expressing effect sizes on a common metric when some studies use transformed scales to express group differences, or use factorial designs or covariance adjustments to obtain a reduced error term. A common metric on which effect sizes may be standardized is described. (Author/RL)

Supposing different values of the standard measurement error, the relation of scale coarseness to the total amount of error is studied on the basis of probability distribution of error. The analyses are performed within two models of error and with two criteria of amount of error. (Editor/SJL)

Discusses problems with scoring reliability of the Vermont Education Department's writing portfolio test, particularly the difficulties teachers face in agreeing upon scoring criteria. (PA)

Because there is no commonly accepted view of what makes for good writing, automated essay scoring (AES) ideally should be able to accommodate different theoretical positions, certainly at the level of state standards but also perhaps among teachers at the classroom level. This paper presents a practical approach and an interactive computer…

Students in an introductory college course (n=275) responded to equivalent 20-item halves of a test under number-right and formula-scoring instructions. Formula scores of those who omitted items overaged about one point lower than their comparable (formula adjusted) scores on the test half administered under number-right instructions. In contrast,…

Numerous techniques are available for determining cutoff scores for distinguishing between proficient and non-proficient examinees. One of the more commonly cited techniques for standard setting is the Nedelsky Method. In response to criticism of this method, Gross (1985) presented a revised Nedelsky technique. However, no research beyond that…

Several studies concerning scoring difficulties on the Wechsler intelligence scales were reviewed. Since scoring of responses on the comprehension, similarities and vocabulary subtests of the Wechsler scales demands judgements by the examiner, the possibility of poor interscorer reliability increases. More thorough scoring standards and revision…

Traditional test scoring does not allow the examination of differences among subjects obtaining identical raw scores on the same test. A configuration scoring paradigm for identical raw scores, which provides for such comparisons, is developed and illustrated. (Author)

The problem of standard setting on language proficiency tests is often approached by the use of norms derived from the group being tested, a process commonly known as "grading on the curve." One particular problem with this ad hoc method of standard setting is that it will usually result in a fluctuating standard dependent on the particular group…

Test scores that are not perfectly reliable cannot be strictly equated unless they are strictly parallel. This fact implies that tau equivalence can be lost if an equipercentile equating is applied to observed scores that are not strictly parallel. Thirty-six simulated data sets are produced to simulate equating tests with different difficulties…

The effects of applying different methods of determining different sets of subscore weights on the composite score ranking of examinees were investigated. Four sets of subscore weights were applied to each of three examination results. The scores were from Advanced Placement (AP) Examinations in History of Art, Spanish Language, and Chemistry. One…

One formulation of confidence scoring requires the examinee to indicate as a number his personal probability of the correctness of each alternative in a multiple-choice test. For this formulation a linear transformation of the logarithm of the correct response is maximized if the examinee accurately reports his personal probability. To equate…

Investigated were the effects of two levels of penalty for incorrect responses on two dependent variables (a measure of risk-taking or confidence, based on nonsense items, and the number of response-attempts to legitimate items) for three treatment groups in a 2x3, multi-response repeated measures, multivariate ANOVA (Analysis of Variance) design.…

A computer program called TEXAN (Textual Analysis of Language Samples) was developed for use in calculating frequency of characters, words, punctuation units, and stylistic variables. Its usefulness in determining readability levels was examined in an analysis of language samples from 20 elementary tradebooks used as supplementary reading…

The answer-until-correct (AUC) procedure requires that examinees respond to a multi-choice item until they answer it correctly. Using a modified version of Horst's model for examinee behavior, this paper compares the effect of guessing on item reliability for the AUC procedure and the zero-one scoring procedure. (Author/CTM)