The Point-Biserial as a Discrimination Index for Distractors in Multiple-Choice Items: Deficiencies in Usage and an Alternative

We show that using the point-biserial as a discrimination index for distractors by differentiating between examinees who chose the distractor and examinees who did not choose the distractor is theoretically wrong and may lead to an incorrect rejection of items. We propose an alternative usage and present empirical evidence for its suitability.