Opposing views are expressed on the value of standardized testing (ST) in public schools. The pro-ST view argues that these tests provide comparative data needed to plan and evaluate schools and individuals. The opposing view focuses on problems of reliability, validity, and bias; and on the damage done to children. (IAH)

A model is proposed for empirically testing the appropriateness of using tetrachoric correlations for a set of dichotomous variables. A testing method is proposed that uses information from trivariate marginal tables. Assumptions are tested with real and simulated data. (SLD)

Explores changes affecting personnel testing and selection during past 20 years, noting that outcomes have often been antithesis of expressed or ostensible reason for change. Contends that some changes, designed to end discrimination, have institutionalized discriminatory practices as consequence of legitimizing methods of preferential treatment…

Reviews previous six papers concerning fairness in employment testing. Contends that papers focus narrowly on legal, theoretical, and societal issues and provide little new information. Argues that industrial psychologists and governmental lawyers are retreating into rigidly defined positions and that the future will be marked by long, costly, and…

In their search for alternatives to standardized assessment, a consortium of Pittsburgh administrators, teachers, and researchers discovered the value of portfolios gathered by artists, musicians, and writers. Now their students gather portfolios consisting of biographies of works, a range of works, and their own reflections. Includes five…

The purpose of this study was to determine the minimal conditions of observation needed to obtain generalizable results across observers, trials, and subjects for the hop, jump, and run at the elementary level. Results are discussed. (JD)

Empirically reviewed Israeli probation officer's recommendation in his investigation report concerning offenders between the ages of 18 and 21, studying factors influencing officer's recommending probation versus refraining from probation recommendation. Tested four models: diagnostic-rehabilitation, prognostic-public defense, organizational, and…

A method of creating linear statistical models that incorporate assumptions implied by hypotheses is described and demonstrated by testing analysis of covariance type hypotheses in the presence of non-parallel regression lines. (SLD)

Actual and theoretical cases are presented in which the Battelle Developmental Inventory was used in infant assessment. Age-related discontinuities reveal that, for children whose chronological age abuts one of the cut-off points between age categories, radically different summary scores can be obtained from one day to the next despite identical…

Describes ways that language test design can be computerized, and illustrates some test construction methods derived from the field of operations research. (CB)

Five counselor performance measures were investigated for their application and use in state-federal vocational rehabilitation agencies using approximately 200 counselors in each of three years. Results suggest that a combination of counselor performance measures may be more effective than any single index. (Author/ABL)

Describes studies designed to examine the instrumentation used to assess both graphing abilities and the impact of microcomputer-based laboratories (MBL) on students' graphing abilities. Results suggest that numerous disparities between the results of multiple-choice and free response instruments. The analysis also revealed that the impact of MBL…

Describes new methods of higher education assessment: outcomes based (at end of course) and value added (multiple measures over time). Reviews other assessment forms--standardized tests, program-based achievement instruments, perception measures--and gives suggestions for home economics student assessment. (SK)

Discusses clinical, efficacy, and validity issues pertaining to the use of computerized IQ testing by comparing computer-based test interpretations (CBTIs) of the Wechsler Intelligence Scale for Children--Revised with reports written by a clinical psychologist. A group of 33 expert clinicians rated the CBTIs significantly higher. (Author/LRW)

Parents of gifted children are urged to understand that there are many different kinds of intelligence and ability tests; tests are designed for different purposes; all tests involve a certain amount of error; and tests only sample one aspect of a child's performance. (DB)