Improving Item Bank Deficits by Modifying Existing Items: A Nudge Versus a Shove

• In Event: Evaulating and Exploring the Nuances of Item and Person Fit

Abstract

Abstract
In computer-based tests (CBT), the ideal distribution of item difficulty reflects the distribution of examinee’s ability, but often tests have a shortage of items toward the harder end of the spectrum. This investigation examines whether existing multiple-choice type items with a known difficulty value could be modified to increase item difficulty in a predictable way. Items from a large item bank were modified by either replacing a distractor (the “nudge” method) or adding a second key to create a new multiple-response type item (a more substantial revision called the “shove” method). Results suggest that the “shove” method was more successful in predictably creating more difficult items.

Authors

• Karen A. Sutherland, Pearson VUE

• John A. Stahl, Pearson VUE

• Ada Woo, National Council of State Boards of Nursing