Distractors, the incorrect responses to an item on a multiple-choice test, should be designed to create confusion in the minds of some students and to permit a competent student to be able to see that they are wrong. It is possible, by close scrutiny, to isolate the sources of this confusion and, by looking at a statistical analysis, to find out which types are most effective. This paper addresses itself to the first of these tasks. An attempt is made to classify distractors by finding out what they are dependent upon, and by inspecting the types of confusion they elicit. Examples are drawn from tests for students of Russian. (Author/CFM)