Using Item Difficulty and Item Position to Measure Test Fatigue

Abstract

With the increase in standardized testing to address accountability issues, educators are becoming more concerned with whether there is a student test fatigue effect on state assessments. If students are not performing as well on items appearing later on tests, efforts to accurately assess performance standards may be compromised. In addition, longer tests may have a differential impact on disadvantaged students. This study compared item difficulties (p-values and b-parameters) from administering the same items in field testing and live testing. Negative changes in item performance based on test position may indicate student fatigue. Contrary to some previous evidence, results showed declines in student performance as items dropped to lower positions on tests.