Poster #59 - Reducing Subjectivity in Value Tables: An Empirically Based Approach

• In Event: 79.052 - MTCC Poster Session 19
  In Poster Session: 79.052-7 - Exploring Issues in Large-Scale Assessments

Abstract

Seven states have used, currently use, or are considering using categorical gain models as part of their state accountability systems in K-12 education. These categorical gain models involve a subjective process of assigning values to a value table to use in conjunction with a transition matrix to make accountability decisions (e.g., “in compliance”) for schools, districts, or the state. Two examples of creating value tables are provided. These subjective decisions are not desirable in a high-stakes accountability system, so a program was developed that uses minimal user input to empirically derive a value table. An example dataset from the Early Child Longitudinal Study – Kindergarten cohort (ECLS-K) is used for purposes of illustration. Applications and uses for the program are discussed.

Author

• Derek Sauder, James Madison University