Using General Outcome Measures to Predict Student Performance on State-Mandated Assessments: An Applied Approach for Establishing Predictive Cutscores

Abstract

Cutscores for reading and math (general outcome measures) to predict passage on New York state-mandated assessments were created by using a freely available Excel workbook. The authors used linear regression to create the cutscores and diagnostic indicators were provided. A rationale and procedure for using this method is outlined. This method allows practitioners to model alternative cutscores and observe effects on correct and incorrect classification of students as passing or failing a high-stakes test. This method of determining cutscores allows school personnel to create defensible and appropriate cutscores on the basis of resources available at the local level.