Dangers in Using the Z Index for Detection of Cheating on Tests

Hynes, Givner, and Patil (1) recently suggested a Z index to be used to statistically test for cheating on a multiple-choice test. It is shown in this work that, although the index may be appropriate for verification of cheating by two previously suspected students, the index should never be used to verify cheating by a single class member when there is no a priori evidence of the identity of the other collaborator. In addition, it is shown that the index should never be used as a screening device to determine whether any cheating has occurred on a test.