Best linear prediction (BLP) and penalized best linear prediction (PBLP) are techniques for combining sources of information to produce task scores, section scores, and composite test scores. The report examines issues to consider in operational implementation of BLP and PBLP in testing programs administered by ETS [Educational Testing Service].

In this technical report, we document the results of a cross-validation study designed to identify optimal cut-scores for the use of the easyCBM[R] mathematics test in the state of Washington. A large sample, randomly split into two groups of roughly equal size, was used for this study. Students' performance classification on the Washington state…

This technical report presents results from a cross-validation study designed to identify optimal cut scores when using easyCBM[R] reading tests in Oregon. The cross-validation study analyzes data from the 2009-2010 academic year for easyCBM[R] reading measures. A sample of approximately 2,000 students per grade, randomly split into two groups of…

In practice, vertical scales have been continually used to measure students' achievement progress across several grade levels and have been considered very challenging psychometric procedures. Recently, such practices have been drawing many criticisms. The major criticisms focus on dimensionality and construct equivalence of the latent trait or…

Educational assessment and, more broadly, educational research in the United States have entered into an era characterized by a dramatic increase in the prevalence and importance of test score use in accountability systems. This volume covers a selection of contemporary issues about testing science and practice that impact the nation's public…

The nature of percentile ranks scores is explored using concrete heuristic examples. It is explained why arithmetic operations require measurement on equal-interval scales, and that percentile ranks are not measured on equal-interval scales and therefore may not be added or averaged. The consequences of inappropriately adding percentile ranks are…

Year-to-year rater variation may result in constructed response (CR) parameter changes, making CR items inappropriate to use in anchor sets for linking or equating. This study demonstrates how rater severity affected the writing and reading scores. Rater adjustments were made to statewide results using an item response theory (IRT) methodology…

This report deals with ways to report basic skills test data which would facilitate the identification of student weaknesses. Under study are the technical aspects and methods associated with the reporting of objective-referenced data. An exploration is then made into the use of patterns of errors in responding to basic skills test items to…

Efforts have been made to increase the dissemination and use of data generated by the National Assessment of Educational Progress (NAEP). Potential users include those concerned with curriculum and methods evaluation, public policymakers, and researchers. NAEP can provide data for curriculum evaluation, including item analysis data which assist in…

The Teacher Evaluation and Assessment Center (TEAC) was established by the Department of Education at the University of South Florida in 1984 to serve the state in the certification of trainers and observers of the Florida Performance Measurement System (FPMS) and to score and report performance evaluations for special programs. This report…