Item Response Theory Analysis of the Western Aphasia Battery

The purpose of this investigation was to improve the validity of the Western Aphasia Battery for measuring aphasia severity and change over time by examining its fit to an item response theory (IRT) measurement model. The advantages of IRT include interval scaling, potential for equating scores across aphasia tests, and more veridical reporting of score reliability. Despite reasonable overall fit to the model, a small number of WAB items demonstrated substantial misfit, suggesting that they do not productively contribute to the measurement of aphasia severity. Potential explanations for this misfit, and implications of IRT for aphasia testing will be discussed.