Drawn from classical test theory. Statistical index (D value) that denotes the difference in proportions of students who get an item right in selected criterion groups.