Evaluation of Flagging Criteria for Anchor Item Stability Analysis

• In Event: Hilton Roundtable Session 4
  In Roundtable Session: Division D Section 1 Roundtable Topic: Applications of Item Response Theory

Abstract

Statewide testing programs commonly use the anchor test design to place student scores from different administrations onto one scale. The d-squared method and robust z statistic are often used to ensure the anchor set is performing similarly across administrations. This simulation study evaluates different criteria of d-squared and robust z for evaluating and selecting the optimal anchor set under Rasch model.

Authors

• Kate Nolan, Curriculum Associates

• Nina Deng, Questar Assessment Inc.