Presents a procedure for standard setting that involves the cluster analysis of test takers to discover examinee groups that are useful for envisioning marginally competent performance or defining borderline or contrasting groups. Illustrates use of the procedure with a statewide mathematics test, and concludes that cluster analysis is useful in…

Provides an overview of the recently developed state accountability system for public elementary and secondary schools in Rhode Island. Discusses accountability systems as technologies of social control and considers the costs, benefits, uncertainties, and necessary tensions associated with accountability systems. (Author/SLD)

Studied the beliefs and practices of 31 teachers about classroom assessment and the Washington state assessment program. Data reveal teachers' preferences for multiple assessments, their objections to the state test for certain groups of students, and their preferences for specialized assessments respectful of student circumstances and readiness.…

Regulations and procedures for the operation of the Virginia General Educational Development (GED) Testing Program are provided. The GED is a means for measuring an individual's educational competency, particularly the major generalizations, ideas, and intellectual skills that are the outcome of four years of high school instruction. Emphasis is…

This brief paper provides an outline for a discussion about the legality of Florida's functional literacy testing program. It defines relevant terms, explains relevant laws, and sets forth the legal basis for the suit and the remedies sought. The discussion is intended to provide a legal and historical framework within which the developments in…

The Arizona Measure of Academic Progress (MAP) was developed using the Stanford 9 Achievement Test (SAT9) scores for the period from Spring 1998 to Spring 1999. The Research and Policy Division of the Arizona Department of Education matched students who took the SAT9 in Spring of 1998 and 1999. On average, 89% of students were matched per grade…

The objectives of this paper are threefold. First, a model is proposed for unifying massive amounts of conceptual and numerical information flowing from the measures of the accountability movement in Ohio and the materials that are publicly available to educators. Second, this model is translated into useable forms of information that help…

This policy brief highlights findings from an annual review of how far states' work on standards has progressed over the years and a determination of how much work remains to be done. Information was collected from state departments of education and interviews with state officials. Since 1995 there has been a steady rise in all aspects of…

This reference guide is designed to give educators and other interested persons a better understanding of the information contained in the score reports from the Washington Assessment of Student Learning, the key measuring stick of the state's school improvement strategy. To do this, the guide is organized around the four basic form types used in…

Reviews the history of initial teacher licensure testing in the United States and summarizes issues and problems, focusing on Massachusetts, a state that has only recently begin initial licensure testing. Provides information on past practices and a table of current state approaches to teacher licensure. (SLD)

The attempt by D.Edwards and C. B. Cummings to adjust state mean Scholastic Aptitude Test Scores for differential participation rates with a "fuzzy truncation model" satisfies three criteria the authors previously defined but falls short for two. Omission of sensitivity studies mars the otherwise exemplary study. (SLD)

A new approach for examining the quality of judgments from standard-setting judges using a Binomial Trials Model (BTM) is presented and illustrated with 26 judges from the Georgia High School Graduation Test. Results suggest that the BTM provides information not available from other methods. (SLD)

This study empirically examined the behaviors of item and person statistics derived from item response theory and classical test theory, focusing on item and person statistics and using a large-scale statewide assessment. Findings show that the person and item statistics from the two measurement frameworks are quite comparable. (SLD)

Investigated how well the generalized and Rasch models described item and test performance across a broad range of mixed-item-format test configurations (six tests from two state proficiency testing programs). Evaluating the impact of model assumptions on the predictions of item and test information permitted a delineation of the implications of…

Studied Kentucky's efforts to include students with disabilities in its statewide assessment, exploring the inclusion achieved, the kinds of accommodations offered, student performance, and the relationship between performance and the use of accommodations. Statewide results show that most students were included, but that some scores may not be…