Three, four, and none of the above options in multiple-choice items

Abstract

High-quality multiple-choice (MC) items are essential for creating efficient, valid assessments. Haladyna, Downing, and Rodriguez (2002) suggested that using plausible distractors is crucial to achieving this goal, although distractor creation can be time-consuming and challenging. Haladyna et al. (2002) provided two related test development guidelines: #18, “Write as many plausible distractors as you can,” and #25, “Use carefully None of the above.” This research aims to test the impact of these two guidelines on item difficulty (p), item discrimination (r), and test reliability for mathematics items empirically. The research findings have revealed that item discrimination and test reliability were not statistically different across MC items with four options, three options, and NOTA options while the means of item difficulty of four-option MC items was not statistically different from those of three-option and NOTA-option MC items. However, the mean of item difficulty of NOTA-option MC items was statistically lower than those of three-option.