Seventeen statistics for measuring the reliability of criterion-referenced tests were critically reviewed. The review was organized into two sections: (1) a discussion of preliminary considerations to provide a foundation for choosing the appropriate category of "reliability" (threshold loss function, squared-error loss-function, or…

Rajaratnam, Cronbach and Gleser's generalizability formula for stratified-parallel tests and Raju's coefficient beta are generalized to estimate the reliability of a composite of criterion-referenced tests, where the parts have different cutting scores. (Author/GK)

Findings support Koppitz's assertion that the total error score for the Bender Gestalt Test is stable and reliable. Working time is a relatively stable dimension of Bender performance, which may be of value in assessment activities. Perseveration and integration should not be used in differential diagnosis. (Author)

For (0, 1) scored multiple-choice tests, a formula giving test reliability as a function of the number of item options is derived, assuming the "knowledge or random guessing model," the parallelism of the new and old tests (apart from the guessing probability), and the assumptions of classical test theory. It is shown that the formula is a more…

Because previous reports have suggested that the lowered validity of tests scored with empirical option weights might be explained by a capitalization of the keying procedures on omitting tendencies, a procedure was devised to key options empirically with a "correction-for-guessing" constraint. (Author)

When multiple-choice tests are scored in the usual manner, giving each correct answer one point, information concerning response patterns is lost. A method for utilizing this information is suggested. An example is presented and compared with two conventional methods of scoring. (Author/JKS)

Fabricated test protocols were used to study how effectively examiners agree in scoring ambiguous WISC-R responses. The results suggest that, even with the improved WISC-R manual, scoring remains a difficult and challenging task. (Author)

A rebuttal of Frary's 1969 article in Educational and Psychological Measurement. (MS)

Adults took the Rod and Frame, Portable Rod and Frame, and Embedded Figures Tests. Absolute and algebraic frame-effect scores were more reliable and valid than rod-effect algebraic scores. Correlations with the Embedded Figures Test were so low that the interchangeability of these field articulation measures is questionable. (Author/RD)

The Serlin-Kaiser procedure is used to complete a principal components solution for scoring weights for all options of a given item. Coefficient alpha is maximized for a given multiple choice test. (Author/GK)

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)

Ordinal response modes for multiple choice tests are those under which the examinee marks one or more choices in an effort to identify the correct choice, or include it in a proper subset of the choices. Two ordinal response modes: answer-until-correct, and Coomb's elimination of choices which examinees identify as wrong, were analyzed for scoring…

Gilman and Ferry have shown that when the student's score on a multiple choice test is the total number of responses necessary to get all items correct, substantial increases in reliability can occur. In contrast, similar procedures giving partial credit on multiple choice items have resulted in relatively small gains in reliability. The analysis…

To date a theoretical basis has not been developed for determining changes in reliability when score points from random guessing are eliminated and those from non-randon guessing are retained. This paper presents a derivation of an expression for the reliability coefficient which displays the effect of deleting score components due to random…

A new formula for the correction for chance success due to guessing was advanced and investigated, the mathematical solution of which has the property of equating the scores of fast and slow examinees of equal ability. (Author/JKS)