A MODIFIED FORM OF THE TRUE‐FALSE TEST

Ever since McCall1 first formally introduced the true-false test, many articles in educational magazines, as well as chapters in pedagogical books, have been devoted to this popular "new type" test. Moreover, a careful examination of the complete bibliography of such testing, prepared by Brinkley,2 reveals that most of the contributions bearing upon the true-false test either describe the test, analyze its merits and shortcomings, or discuss the statistical considerations relative to the guessing factor as it affects the scoring. To date few attempts to modify the form of the true-false test so as to eliminate objections which are generally recognized as inherent in its present form have been recorded ; but expressions of a desire for its improvement are not infrequently found. Barthelmess, for example, points out the weakness of the true-false "as an instrument for diagnosing special individual difficulties," and says further, "if a test could be devised which would have all the advantages of the true-false and correct for this disadvantage, it would be heartily wel comed."3 A few modifications of the test, however, have been sug gested. Knight requires his students to mark the test questions "T," "F," and "U" (undecided) and gives them time only to read each statement and make an instantaneous decision. An ad vantage of this modification, he thinks, is that it "discourages the proclivity to borrow from one's neighbors."4 Farwell gives