A Bayesian framework for making mastery/nonmastery decisions based on multivariate test data is described. Overall, mastery is granted if the posterion expected loss associated with such action is smaller than the one incurred by denying mastery. (Author/JKS)

Approaches to the determination of cut-scores have used threshold, normal ogive, linear and discrete utility functions. These approaches are examined by investigating conditions on the posterior, likelihood and utility functions required for setting cut-scores in a Bayesian approach. (Author/JKS)

The author describes an assessment model which applies a databased, criterion-referenced formative testing approach to define learning problems, determine teaching strategies, and evaluate progress in learning disabled students. Three case studies illustrate the procedures for implementing the approach. (SB)

The author argues that competency-based education does not fit all vocational education applications and should be used in moderation. He also discusses vocational education's focus on actual occupational tasks and the use of criterion-referenced measurement of performance. (CT)

Describes a procedure for using the computer to assist in evaluating the progress of students on pretests, unit tests, posttests or a combination of tests. The use of computers to evaluate cognitive objectives of a course is examined. Twenty-four references are listed. (MER)

Describes second language testing methods including discrete-point tests, integrative tests, criterion-referenced testing, FSI test of oral proficiency, tests of functional ability. No way has yet been found to reliably measure communicative competence in second languages. (BK)

Briefly discusses the phases and components of the IPISD (Interservice Procedures for Instructional Systems Development), reviews three critical points (initial analysis, determining training objectives, criterion referenced test development), and provides recommendations to assist the performance technologist in playing the difficult roles of…

A nonrandomized minimax solution is presented for passing scores on mastery tests using the binomial error model. The computation does not require prior knowledge regarding an individual examinee or group test data for a population of examinees. A scheme which allows for correction for guessing is also described. (Author/JKS)

Taylor and Russell produced a set of tables which gives the proportion of successful cases when a given cutoff score or a predictor variable and the correlation between the predictor variable and the criterion are known. This paper extends this procedure to the case where parameters are not known. (Author/JKS)

Briefly describes a program used in the Hazelwood School District in Missouri. (JM)

Thirty states which administer criterion-referenced tests to students with visual impairments were surveyed to determine types of accommodations they allowed in their testing programs. The survey addressed aspects of the test setting, test scheduling, test directions, test format, answer mode, and mechanical and nonmechanical aids. (Author/JDD)

Presents simplified methods for deriving estimates of the consistency of criterion-referenced, English-as-a-Second-Language tests, including (1) the threshold loss agreement approach using agreement or kappa coefficients, (2) the squared-error loss agreement approach using the phi(lambda) dependability approach, and (3) the domain score…

Reflection by 15 nurses in the Regents College Nursing Program, an external degree program based on four criterion-referenced tests, uncovered motivations for selecting the program, clarified such needs as support networks and learning resources, and revealed a pervasive pattern of stress that affected exam preparation and performance. (SK)

Proposes a hierarchical item response theory (IRT) model for mastery classification in criterion-referenced measurement. In this model, items measuring the same criterion are grouped, and a difficulty and discrimination parameter for the criterion is estimated on the same scale as the person and item parameters. Illustrates the model with a test…