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)

An alternative interpretation of Livingston's reliability coefficient (see TM 500 487) is based on the notion of the relation of the size of the reliability coefficient to the range of talent. (Author/CK)

Proposes that, while there may be real sex differences in vocational interests, Gottfredson's reanalysis of Self-Directed Search predictive data provides no support for the counseling use of sex-divergent raw interest scores. (RC)

Three methods are presented for interpolating fall norms from data derived for tests administered in spring. One method used the midpoint between two spring administrations, the second adjusted for date of fall testing, while the third method used experimental data that showed slower growth rates during the summer months. (CTM)

The author defines criterion-referenced (CRT) and norm-referenced (NRT) tests, considers current objections to each, and distinguishes their appropriate uses: CRTs for pupil diagnosis and instruction and NRTs for surveying achievement in a long-term evaluation context. Part of a theme issue on student assessment. (SJL)

Scores on the male and female scales of the Strong-Campbell Interest Inventory were compared for college students (N=525) to evaluate the claim that male scales should be used with female counselees. Both sexes scored higher on opposite-sex scales. Score relationships support using only same-sex score interpretations with counselees. (Author)

The Faculty Beliefs about Grades Inventory, a three-scale instrument, was developed to measure faculty beliefs about norm- and criterion-referenced grading and attitudes toward the sorting and selection functions of grades. Results with 243 college faculty members show high negative correlation between attitudes about norm-referenced and…

Studied longitudinal changes in performance at both the student and school district level in major content areas of a widely used norm-referenced grade-level testing program. Used data from grades 3 to 4 and from 7 to 8 of the Iowa Tests of Basic Skills (in Iowa). Reports descriptive statistics and empirical norms and reliability estimates for…

Fifty-six school psychologists reviewed a hypothetical case study designed to investigate their interpretations of different norm-referenced test scores. Subjects were more likely to recommend special education programs for normal students when percentile scores were used. School psychologists need to become more familiar with the properties of…

Six methods of scoring multiple true-false items were compared in terms of reliabilities, difficulties, and discrimination. Results suggest that, for norm-referenced score interpretations, there is insufficient evidence to support any one of the methods as superior. For criterion-referenced score interpretations, effects of scoring method must be…

The feasibility of combining criterion-referenced and norm-referenced tests into a school achievement test score to be used for school effectiveness classification was studied with 361 elementary schools across 2 years and across a subsample of 264 schools. Results support combining scores of different tests to measure school effectiveness. (SLD)

Two criteria defined in previous research that can be used to evaluate the validity of normative data provided for customized tests are discussed. Results of an exploratory investigation of the validity of such data for about 2,500 fifth graders in a 1989 study are reported. (SLD)

Study determined test-retest reliability for 1-mile, three-quarter-mile, and half-mile distance run/walk tests for K-4 students. Students completed testing in various distance conditions. The distance run proved reliable with appropriate age/distance adjustments. Test scores were generally reliable in classifying students with regard to meeting…

Application of a knowledge-based standard in evaluating a written certification examination developed by the American Board of Anesthesiology established a standard of 57 percent correct over two years' examinations. This process is recommended for developing mastery-based (rather than normative-based) success criteria for evaluation of medical…

This article describes errors resulting from use of standardized language tests using a "normal" sample as the normative comparison group. Resulting errors include, among others, identifying normal children as language impaired, providing misleading profiles of verbal and nonverbal performance, and inability to determine impairment severity.…