The criterion-related validity evidence and other psychometric characteristics of multiple-choice and multiple true-false (MTF) items in medical specialty certification examinations were compared using results from 21,346 candidates. Advantages of MTF items and implications for test construction are discussed. (SLD)

Critics voice concern over the validity of multiple-choice tests and suggest alternatives which focus teaching more on thinking skills and less on rote learning. Essay tests are an authentic form of testing currently used from middle school through college. The article recommends training teachers in the area of testing. (SM)

This study is part of a larger one examining the comparability of the First Certificate in English and the Test of English as a Foreign Language. The general assumption of unidimensionality and goodness-of-fit were tested. Findings raise questions about the consequences of rejecting or retaining misfitting items. (60 references) (LB)

The design of a valid and reliable test of test-wiseness is reported: a 33-item multiple-choice instrument with 4 subscales trialed with several groups of English-as-a-Second-Language students. Findings indicate differential skills in test-taking; some learner scores are influenced by skills that are not the focus of the test. (13 references)…

Two studies are reported that examined translation from second language to first language as a testing procedure. Results suggest that the widespread rejection of translation as a language testing procedure by teachers and testers is probably not warranted on psychometric grounds, but that it should be used with extreme care. (10 references)…

This essay notes the complexities inherent in classifying pervasive developmental disorders (PDD), especially when a single classification scheme is to be used for clinical, research, and administrative purposes. The paper considers principles of classification, symptomatology, the neuropsychiatric interface, the value of a multiaxial approach,…

Forty-three adolescents with autism and developmental impairments were assessed with the Autism Diagnostic Interview (ADI), DSM-III-R (Diagnostic and Statistical Manual III, Revised) criteria, and the clinician's diagnosis. DSM-III-R criteria for autism exhibited low specificity and agreed poorly with the other two definitions, whereas the ADI was…

Forty-eight teachers and 41 program administrators working with learners with severe disabilities completed the Program Quality Indicator checklist. Both groups consistently rated the majority of the 123 criteria as important or very important. The two dimensions considered most important were also ranked most important by six professional groups…

The Stanford-Binet IV is compared to the original version and criticized for having less power to measure the high end of intelligence and for having norms that discriminate against gifted students. Strengths of the Stanford-Binet L-M are pointed out, and use of both scales for different purposes is recommended. (JDD)

Variables that constitute translation ability are discussed, based on a two-year development and validation study of job-related tests of translation ability for the Federal Bureau of Investigation. The project involved the development of two parallel forms of the Spanish into English Verbatim Translation Exam (SEVTE). (five references) (LB)

The Miller Assessment for Preschoolers, intended for children ages 2-5, is designed as both a screening and a diagnostic assessment, measuring neuromaturational variables, coordination, language, memory, problem solving, and visual perception. This review examines test administration, summation of data, standardization, reliability, and validity,…

This critique of facilitated communication as an aid for individuals with autism reviews its use in Australia and discusses such issues as the high degree of physical intervention provided by the assistant and the unlikely language content exhibited. An editorial comment deplores the media hype promoting facilitated communication but admits its…

The psychometric characteristics of the Job Descriptive Index and other inventories of job satisfaction are examined across a sample of validation studies. The construct validity of these inventories is modest, but this problem can be corrected through proper application of structural equation models. (SLD)

Predictive validity of preadmissions measures may be understated because of correctable defects in freshman year and cumulative high school grade point averages (GPAs). A study used item response theory (IRT) to develop a more reliable measure of performance and test it using Stanford University data. Results showed increased predictability.…

This article concludes that the Stanford-Binet Intelligence Scale: Fourth Edition is poorly suited for assessment of children less than five years old with possible mild mental retardation and for persons of any age with possible severe mental retardation. The continued use of Form L-M is recommended for these populations. (Author/DB)