ERIC Number: ED480877
Record Type: Non-Journal
Publication Date: 2003-Mar-27
Reference Count: N/A
The Performance Levels and Associated Cut Scores on the Pennsylvania System of School Assessment Mathematics and Reading Tests: A Critical Analysis.
Zwerling, Harris L.
In the context of controversy over the use of high stakes testing, the Pennsylvania State Education Association (PSEA) asked for an evaluation of the performance levels and cut score of the Pennsylvania System of School Assessment (PSSA) mathematics and reading tests. While awaiting technical documentation from the Pennsylvania Department of Education (PDE), the PSEA began an evaluation of its own. Ten Pennsylvania school districts were asked to participate in the study, and four agreed to do so. Part 1 of this report discusses the technical information provided by the PDE to the public and before the State Board of Education and that provided in response to the PSEA Right To Know requirements. It assess the adequacy of the cut score process. Part 2 contains the study jointly conducted by PSEA and the Pennsylvania Association of School Administrators (PASA) of the relationship between PSSA performance levels and performance on commercial standardized tests. This study was conducted to provide a gauge of the external validity of the cut scores. The final section contains conclusions and recommendations. Instead of selecting the more fully documented and carefully executed (bookmark) method for establishing cut scores, the Pennsylvania Secretary of Education averages the cut scores established using two methods, a bookmark and a borderline group method, and then increased the cut scores by a quarter standard error. This data manipulation resulted in tens of thousands of students falling into lower performance categories. For all six PSSA tests, the cut scores approved by the state, when compared to the bookmark study cut scores, decrease the number of students achieving proficient or better by more than 25,000. The PDE cut scores increase the number of students failing by more than 42,000. Findings show that many students failing the PSSAs do quite well on other commercial tests. It is also noted that below-basic PSSA performers averaged slightly higher Scholastic Assessment Test scores than those classified as basic by the PSSA, providing more evidence that the failure cut score at grade 11 requires re-examination. An appendix contains a request for test information, and the other three contain data tables. (Contains 20 graphs, 3 tables, and 21 references.) (SLD)
Publication Type: Reports - Research
Education Level: N/A
Authoring Institution: N/A
Identifiers - Location: Pennsylvania