Two rating instruments for evaluating the performance of men in Navy jobs by supervisors were developed based on the need to know how effectively men of varying ability perform different tasks. Specific Navy jobs were selected for the development of the performance instruments according to the following job factors: Highly populated,…

Discusses the issue that teachers in the Caribbean neglect the education of their pupils to prepare them for examinations, and offers an outline of the elements involved in the development of a syllabus that will educate pupils in the fullest sense, and from which the examination will follow. (Author/AM)

Examines the problems of constructing tests and shows how the rules of transformation generative grammar can be used to produce easily understood test questions. (RB)

This article presents suggestions for improvement in constructing better essay questions and for teaching students how to organize and write essay answers. (Author/ND)

Two procedures for the display of item analysis statistics are described. One procedure allows for investigation of difficulty; the second plots item difficulty against item discrimination. (Author/JKS)

Describes a computer-managed instructional system used to formulate, print, and evaluate true-false questions for testing purposes. The design of the system and its application in medical and nuclear engineering courses in two Spanish institutions of higher learning are detailed. (RAO)

Compares four different scoring systems in a study of 103 subjects answering a 20 item multiple choice test. Three of the four systems uses a weighted response technique, giving credit for partial knowledge. This seemed to have a significant effect on final grades and gave a more accurate grade representation. (MA)

The root of problems in standardized testing is the test makers' goal of lining up children along a single line from highest-scoring to lowest-scoring individual. Standardized tests reflect no sense, no reality, and no logic. (MJB)

Several Bayesian approaches to the simultaneous estimation of the means of k binomial populations are discussed. This has particular applicability to criterion-referenced or mastery testing. (Author/JKS)

Validation studies of the Course-Faculty Evaluation Instrument (CFI) were analyzed, using a multitrait-multimethod (MTMM) strategy. Using an analysis of variance framework to summarize MTMM matrices, the CFI demonstrated greater discriminant validity using more dimensions (traits) and had a lower error component than the Courseg Scales

Describes the construction of a test item file for an Introductory Astronomy course. The questions are typed on cards, arranged by a content categorization scheme for quick retrieval, and classified according to behavioral objectives. (GA)

Investigates empirically, for a sample of university examinations in engineering, the influence of examination length on reliability and discrimination. (Author)

Measures of consistency and completeness of order relationships derived from test data such as Guttman scales are proposed. The measures are generalized to apply to incomplete data such as data from tailored testing. (Author/JKS)

Urry's procedure for approximating latent trait test models is shown to tend to underestimate item discriminatory power and overestimate item difficulty. A method for correcting these biases is provided, and implications of the procedures are discussed. (Author/JKS)

TAILOR, a FORTRAN computer program for tailored testing, is described. The procedure for a joint ordering of persons and items with no pretesting as the basis for the tailored test is given, and a brief discussion of the computer program is included. (Author/JKS)