A Meta-Analytic Review of Item Discrimination and Difficulty in Multiple-Choice Items Using "None-Of-The-Above"

Abstract

The research on using "none-of-the-above" (N-O-T-A) as a test item option continues to find contradictory results on item difficulty and item discrimination. The argument for not using N-O-T-A is that this option is considered more difficult and less discriminating than items with one correct format (regular).The purpose of this study was to conduct a meta-analysis on the difficulty and discrimination of the N-O-T-A test option. A total of twelve articles yielding 20 effect sizes was examined for difficulty, and a total of seven studies yielding eleven effect sizes was examined for discrimination. The meta-analysis indicated nonsignificant effect sizes of .01 for discrimination and -.17 for item difficulty. These findings indicate that using "none-of-the-above"as a test item option does not result in items of lesser quality than in items not using this option.