Describes a dozen ways in which reading tests are at variance in their demands with the demands of reading as it occurs in everyday life. Suggests that these sources of variance render reading tests considerably less valid as measures of real world reading behavior than most people want to believe. (RS)

A Bayesian alternative to interpretations based on classical reliability theory is presented. Procedures are detailed for calculation of a posterior score and credible interval with joint consideration of item sample and occasion error. (Author/SLD)

An approximate statistical test is derived for the hypothesis that the intraclass reliability coefficients associated with two measurement procedures are equal. Control of Type 1 error is investigated by comparing empirical sampling distributions of the test statistic with its derived theoretical distribution. A numerical illustration is…

This study evaluated the technical qualities of a naturalistic test of reading decoding skills, the"Diagnostic Names Task." Evaluation of the measure with 25 elementary grade disabled readers examined subtest reliability, sensitivity to change resulting from instruction, meaningful clustering of subskills, and criterion-related validity.…

Reliability and accuracy of peer ratings by 32, 28, 33 general surgery residents over 3 years were examined. Peer ratings were found highly reliable, with high level of test-retest reliability replicated across three years. Halo effects appear to pose greatest threat to rater accuracy, though chief residents tended to exhibit less halo effect than…

Because of the importance of single-subject studies in mental retardation research, a meta-analysis of the literature examining interrater agreement for visual analysis of single-subject data was conducted. Analysis revealed an overall itinerant agreement value of 0.58 for 14 studies representing 789 raters. Effects of data type, design type,…

Discusses online search techniques that can be used to identify inaccuracies and inconsistencies in records of online and CD-ROM databases. Methods that can be used to sample a database for errors in descriptor, name, or coded fields are described, including browsing, matching, looking up cross-references, and searching for implausible record…

Confirmatory factor analysis using the LISREL computer program was performed on test scores from 500 workshop participants (adults) taking the Myers-Briggs Type Indicator--Expanded Analysis. Results, cross-validated with 500 other scores, support 4 factors as well-defined constructs. Improvements are suggested for subscale reliability. (SLD)

The formula developed by R. Levine (1955) for equating unequally reliable tests is described. The formula can be interpreted as a method of moments estimate of an equating function that results in first order equity of the equated test score under a classical congeneric model. (TJH)

Sensitivity and specificity of static acoustic admittance (SAA) and tympanometric width (TW) were assessed for 116 pediatric patients. A statistically significant change in SAA was observed in immediate repeat measures. No significant change across the two sequential measures was observed for TW, peak pressure, or the estimate of ear canal volume.…

Examined internal consistency reliability of Kaufman Assessment Battery for Children (K-ABC) for 311 Black and 1,450 White students between ages of 2.5 and 12.5 years. Differences between various K-ABC composite, subscale, and subtest reliability coefficients for two groups were nonsignificant; reliability coefficients across comparisons were…

Discussion of indexing consistency and its possible effect on information retrieval focuses on a study that compared indexing (or cataloging) consistency between Library of Congress catalogers and British Library catalogers assigning Library of Congress Subject Headings. The average indexing consistency value is calculated for 82 titles in library…

This replication study of time-interval judgments of stuttering found higher interjudge agreement than previously reported for event-based analyses of stuttering judgments or time-interval analyses of event judgments. Judges with high intrajudge agreement levels also showed higher interjudge agreement levels than did judges with low intrajudge…

This study, involving parents of 165 children under 6-years old, revealed internal consistency and test-retest reliability of 9 of the 11 dimensions constituting the Parent Interview for Autism. Six of the test dimensions differentiated young children with autism from a developmentally comparable group with mental retardation. (Author/JDD)

Holistic scoring enables the evaluation of writing skills based on general impressions of content and style. An experiment in an accounting class shows how it can be applied successfully with a high degree of reliability. (SK)