Lissette A. Tolentino
Anne Corinne Huggins-Manley
University of Florida
Abstract
High-stakes testing in education often requires the use of cut scores to report achievement. In Florida, cut scores are used to establish different levels of proficiency. Although the Florida Standards Assessments (FSA) reports the accuracy rates for cut scores, it does not report classification consistency, nor does it report information on the alignment between the high-stakes cut scores and variations in classification quality across a range of possible cut scores. Our purpose is to perform a case study evaluating the alignment between marginal classification accuracy and consistency rates across the ability continuum to cut point locations for high-stakes cut scores, and to demonstrate the practical utility of this cut score evaluation method that was proposed by Wyse and Babcock (2016). We achieved this purpose through the use of a large set of simulated test data samples generated from FSA item and person parameter estimates.
Citation
Tolentino, L. A., & Huggins-Manley, A. C. (2019). Examining Alignment of Classification Quality to High-Stakes Test Decisions in Florida. Florida Journal of Educational Research, 57(4), 1-14.
Download:
Tolentino574.pdf (1028 downloads )