Research on the validity of student rating scales of teacher performance is inconsistent, and the issue of why the scales are used needs examination. Until their criterion validity is raised to correlations of .7 to .8 or construct validity can be established, results should not be used in making tenure decisions. (MSE)

Presents approaches to test development analysis, reliability, and validity based on criterion-referenced measurement principles and compares them with norm-referenced approaches in terms of the types of decisions that result from either approach. This is done by using data from an English-as-a-second-language achievement testing project at the…

A study of persistence/dropout among open admissions college students found: (1) accurate predictions cannot be made for individual students at the time of matriculation; and (2) it is possible to predict that about 80 percent of future groups will fall in the persist category after two semesters, 51 percent after four semesters. (CMG)

The Figural, Symbolic, and Semantic dimensions or the Structure of Intellect Learning Abilities (SOI-LA) tests have been employed as prescriptive tools for matching students to appropriate methods of reading instruction. A multiple-group factor analysis revealed strong evidence of their construct validity in a normative sample of 364 second…

Twenty-one studies that reported mean score differences between the Woodcock-Johnson Tests of Cognitive Ability (WJTCA) and the Wechsler Intelligence Scale for Children-Revised (WISC-R) Full Scale are summarized. The differences are found to be minimal and are attributed to data bias and WJTCA's higher correlation with achievement. (EGS)

PLANA is a training program for subject matter analysis and construction of instructional objectives. This evaluation compared the performance of PLANA-trained and control-group teachers as to their precision in defining teaching objectives and their corresponding instructional tasks and test items. (BW)

Findings and research designs used to study university students' evaluations of teaching effectiveness are reviewed. A construct validation approach which recognizes the multidimensionality of both effective teaching and students' evaluations is recommended for further research. (BS)

Research focusing on the construct validation of self-concept (SC) within an educational framework is reviewed. Studies are categorized as within-network research and between-network research and further classified by methodological procedures. Although relationships among self-concept facets and academic achievement have been found, an overall SC…

Notes that, while the test is not recommended for children at the lower and extreme upper levels, it does provide a valid and reliable measure of some of the abilities that may be involved in the reading behaviors of five- and six-year-old children. (FL)

Bowdoin College has allowed applicants to submit or withhold SAT scores since 1970. Admissions operations under this policy are reviewed. The validity of the SAT is evaluated, and the academic records of matriculated submitters are compared with the withholders. (Author/MLW)

This study analyzed the Tennessee Self Concept Scale (TSCS) scores from three subject groups (one clinical and two college samples). Five separate item pool arrangements were organized from the TSCS to assess its validity and related issues. The utility of the TSCS in applied settings is discussed. (Author/BS)

Discusses methods of sampling the best information from instruction/training developers/candidates for professional certification and examines the problems of interpreting that information and making classification decisions. Assessment strategies including criterion-referenced, multiple-choice, short answer, and essay questions, and portfolio…

Concludes that the Brigance Screen is a well-organized criterion-referenced test designed to assist in the early identification of individuals who need further testing. (FL)

Reports that reading instructional level scores of a teacher constructed informal reading inventory, the commercially prepared Basic Reading Inventory, the Metropolitan Achievement Test, and students' actual level of placement in books are roughly comparable but that the Wide Range Achievement Test reading subtest places children much higher. (FL)

Reviews literature on children up to age one to determine whether heart rate (HR) is a reliable, stable, and valid measure. Considers factors influencing resting HR and the Law of Initial Values, discusses information about response to stimulation and infant perception and affect obtained by using HR measures, and describes HR use in risk…