Single-Scorer School Grading Formula. Research Brief. Volume 1001Report as inadecuate

Single-Scorer School Grading Formula. Research Brief. Volume 1001 - Download this document for free, or read online. Document in PDF available to download.

Research Services, Miami-Dade County Public Schools

Due to budgetary concerns, the FDOE restructured the Writing portion of the FCAT assessment program for 2010. Among the changes was a difference in how the writing essays were scored. In the past, two people independently judged each essay and each assigned a score from 1 to 6. In the event of a disagreement between judges, the average of the scores was assigned. Thus, "half-value" final scores like 2.5 and 3.5 were possible results. When it came to grading schools, the convention was to have the percent of students scoring 3.5 or higher constitute the writing component of the overall point total. However, in 2010 only a single judge would score each essay. Scores between integer values (2.5, 3.5, etc.) would no longer be possible. Since a score of 3.5 could not occur, it would not be possible to summarize a school's performance in an equivalent manner comparable to the "percent scoring 3.5 or higher" standard of the past. To keep the school summary scores as alike in meaning as possible, some accommodation to the school grading methodology would have to be introduced. Since a numerical school summary defined by the "percent scoring 3.5 or higher" would not be strictly possible, the initial solution considered was to simply substitute the average of the "percent scoring 4 or higher" and the "percent scoring 3 or higher." On the surface, this seems to make sense; after all, the average of 3 and 4 is 3.5. However, while this proposal superficially seems straightforward and fair, there is a subtle source of systematic bias hidden in that kind of computation. The purpose of this paper is to explain the nature of the calculation error and suggest an alternative procedure that would provide a more accurate estimate consistent with the "3.5 or higher" traditional approach.

Descriptors: School Effectiveness, Grading, Measurement Techniques, Computation, Mathematical Formulas, Writing Tests, Achievement Tests, Scoring

Research Services, Miami-Dade County Public Schools. 1450 NE Second Avenue, Miami, FL 33132. Tel: 305-995-1000; Fax: 305-995-7521; Web site:

Author: Froman, Terry; Shneyderman, Aleksandr



Related documents