The Whole Is Greater Than the Sum of Its Parts: Research on Aggregated Test Scores

• In Event: Roundtable Session 117

Session Submission Type: Roundtable Session

Abstract

In educational practice we often administer several tests to assess a single student. For example, when several tests are combined into one diploma-decision. Depending on the context, the content and accuracy of the combined components may differ. In some instances the components are designed to accurately measure various constructs and in other instances relatively inaccurate measurements of the same construct are aggregated. The purpose of this session is to combine different studies that address methods that can be applied when aggregating test scores. The topics addressed are accuracy and validity of a decision based on aggregated tests, profile analysis for reporting on subtests, and dynamic models to aggregate repeated measurements of a single construct.

Sub Unit

• Division D - Measurement and Research Methodologies / Section 1: Educational Measurement, Psychometrics, and Assessment

Chair

• Anton Beguin, Cito

Papers

• Reliability of Aggregation of Results on Multiple Assessments: Which Role Can Retests Play? - Robert Zwitser, Cito

• A Valid Decision Based on Multiple Tests: Validation of Competence Assessment Programs - Saskia Wools, Cito

• Profile Analysis: A Generalization of Differential Item Functioning Analysis - Norman D. Verhelst, Eurometrics