Studied interviewer reliability, validity, and strategy for information integration. Candidates (N=412) for selection to a military division were interviewed and assessed. Results indicated that interviewers functioned in a similar fashion. Analyses of individual interviewers indicated higher reliability and individual differences among…

The Peabody Picture Vocabulary Test-Revised (Dunn and Dunn) is described as a convenient, quick test, possessing improvements over the original. It measures a subject's receptive (hearing) vocabulary for Standard American English. However, the validity information for the test is less than adequate, since no validity studies are presented for it.…

The psychological and linguistic foundations of the cloze procedure are outlined, and standard approaches and terms used are reviewed. Notes are given on cloze in testing, in relation to language structure, and as a teaching device, focusing on the manipulation of the technique for a variety of purposes and situations. (MSE)

This study was concerned with the reliability and validity of subjective judgments about five characteristics of multiple-choice test items from an introductory college-level astronomy test: (1) item difficulty, (2) language complexity, (3) content importance or relevance, (4) response set convergence, and (5) process complexity. (Author)

Describes a variety of ways to create cloze blanks and discusses how the reader's task differs in the various types of exercises. (FL)

In a study of tests on diagnosis it was found that confidence measures of knowledge adjusted for reasoning were more reliable than conventional measures of knowledge, supporting the idea that knowledge enhances the ability to recognize the diagnostic entities, and realism is associated with fewer premature closures. (MSE)

In a study of the vocabulary proficiency of some South African ESL teacher trainees, the General Service List of English Words' validity was evaluated. It was found that mastery of this list would meet most of the vocabulary needs of the test group. Recommendations are made for practical uses of word counts. (MSE)

A new method was developed to assist in the selection of a test length by utilizing computer simulation procedures and item response theory. A demonstration of the method presents results which address the influences of item pool heterogeneity matched to the objectives of interest and the method of item selection. (Author/PN)

Two studies of the oral secondary school leaving examinations in Germany indicated variances in ratings made by different teachers. The differences were significantly influenced by speed of speaking, background information about past achievement, and contrast of performance related to sequence of presentation. (Author/MJL)

Uses a scale to analyze three tests used for assessing community college students' reading levels: The Stanford Diagnostic Reading Test (SDRT), the Nelson-Denny Reading Test (NDRT), and the Iowa Silent Reading Tests (ISRT). Judges the ISRT to be the best and the NDRT to be unacceptable according to the criteria of the scale. (FL)

Discusses the roles played by student evaluations of teacher performance in the instructional design process; reviews issues related to their use; and outlines suggestions for the construction, administration, and analysis of the questionnaire data by the instructional development consultant. (Author/JJD)

Addressing Glass' argument (EJ 198 842) that a lack of interrelater reliability is an inherent deficiency in the Nedelsky technique, poor rater training and the need for a group decision procedure are presented as standard setting problems. (CM)

An observation instrument designed to be applicable across teaching situations (both grade levels and educational strategies) is presented with evidence regarding its efficiency and generalizability. The dependability of scores yielded is assessed by a generalizability study, and recommendations concerning number and item types in the measure are…

Findings regarding formats and scoring formulas for multiple-choice test items with more than one correct response are presented. Strong cluing effects in the Type K format, increasing the correct score percentage and reducing test reliability, recommend using the Type X format. Alternative scoring methods are discussed. (Author/CM)

A critique of a variety of formats used in combined-response test items (those in which the respondent must choose the correct combination of options: a and b, all of the above, etc.) illustrates why this kind of testing is inherently flawed and should not be used in optometry examinations. (MSE)