This paper follows a line of logical argumentation to claim that what Samuel Messick conceptualized about construct validation has probably been misunderstood by some educational policy makers, practicing educators, and classroom teachers. It argues that, while Messick's unified theory of test validation aimed at (a) warning educational…

The gold standard in making causal inference on program effects is a randomized trial. Most randomization designs in education randomize classrooms or schools rather than individual students. Such "clustered randomization" designs have one principal drawback: They tend to have limited statistical power or precision. This study aims to…

Setting a cutting score is a key problem in criterion-referenced measurement which is discussed within a decision theoretic approach when just one student is considered. A minimum information solution is given and compared with approaches when there is information about a group of students. Formulas illustrate the discussion. (CM)

The increasing relation between domain-referenced measurement and its implications in instruction are discussed with the argument that construction methods available do not guarantee content validity which needs more attention to be effective. Two domain-referenced testing procedures illustrate construction and validity problems. (CM)

Three item selection methods for criterion-referenced tests are examined: the classical theory of item difficulty and item-test correlation; the latent trait theory of item characteristic curves; and a decision-theoretic approach for optimal item selection. Item contribution to the standardized expected utility of mastery testing is discussed. (CM)

Most educational measurement texts distinguish between norm-referenced (NR), or relative, methods of assigning letter grades to objective test scores, and criterion-referenced (CR), or absolute, methods. Both NR and CR approaches have serious limitations in typical classroom situations, and neither approach, in its pure form, may be entirely…

Norm-referenced tests (NRT) help compare the performance of one student with the performances of a large group of students, while criterion-referenced tests (CRT) focus on "what test takers can do and what they know, not how they compare to others" (Anastasi, 1988). Both types of test can be standardized so that scores can be interpreted the same…

This report is designed to provide numerical data about student performance on the curriculum of the Boston (Massachusetts) Public Schools (BPS) as measured by curriculum referenced tests (CRTs). BPS Department of Educational Testing analyzed data from the 1986 CRT final exams in: reading and language arts (grades K-8), high school English (six…

This article addresses the capacity of high stakes tests to measure the most significant kinds of learning. It begins by examining a set of philosophical arguments pertaining to construct validity and alleged conceptual obstacles to attributing specific knowledge and skills to learners. The arguments invoke philosophical doctrines of holism and…

The effectiveness of the Elementary and Secondary Education Act Title I reading and mathematics laboratories were evaluated to assess the improvement of underachieving elementary school students' academic skills. Laboratory students were selected on the basis of academic (rather than socioeconomic) criteria in grades 2 through 5; the control…

The Texas Assessment of Minimum Skills (TEAMS) is a mandated criterion-referenced test administered to students in grades 1, 3, 5, 7, 9, and 11 in Texas public schools. This report by the Austin Independent School District (AISD) contains an executive summary of TEAMS results, an analysis of performance, and attachments. Among the major findings…

Discusses the movement toward authentic assessment, also called direct or performance assessment, as an alternative to multiple-choice, standardized, norm-referenced testing. Authentic testing involves assessment tasks that are real instances of extended criterion performances rather than proxies of actual learning goals. Questions use of…

Intended both to evaluate the California testing program's success in assessing student achievement in the basic skills and to suggest possible improvements in the program, this report includes an executive summary evaluating current testing procedures, describing experimental studies on new ways of testing and reporting testing information, and…

In fall 1986, a study was conducted at Florence-Darlington Technical College (South Carolina) to measure the reliability and validity of the TEC-MAT, a newly developed entrance test. The TEC-MAT instrument, a criterion-referenced test consisting of 100 items assessing numerical skills, was field tested among a group of entry-level freshmen already…

This review covers a time span from 1922 to 1980. It examines the link that exists between assessment and teaching. After discussing the milieu in which the two take place, the remainder of the review addresses four questions: (1) how do teachers use test information for instructional decisions?; (2) what has research indicated about the linkage…