Herbert A. Toops

Simplified formulas for item selection and construction

The formula for the Pearson correlation coefficient of a dichotomous variable with a multiple-categoried variable is simplified for computational purposes by effecting in the multiple-categoried variable two types of arbitrary distributions: (1) rectangular and (2) proportional to binomial expansion coefficients. The formulas which result are convenient for the selection of test items and are applicable to the objective estimation of the comparative merits of the alternatives in multiple-choice test items. It is shown that the authoritative answer should have a high positive criterion coefficient, while the omissions and several wrong-answer alternatives should each have low (algebraic) negative criterion coefficients.

Philosophy and Practice of Personnel Selection

BY definition selection implies more candidates than jobs, a choosing of the most fit. In boom years and wars selection wanes; it waxes in depressions and peace. Following the war it will become important again. It has become obvious that much of our material progress is due, on the one hand, to a very few expert people who are able to invent such things as B-29’s, radar, dehydration, and penicillin; and equally, on the other hand, t o a multitude of Joe’s, Bill’s and Sally’s whose skill of hand, keenness of eye and sureness of touch, in small things, just as surely is an expertness of its own. Some kinds of people do each of these respective kinds of work better than others. Subdivide and specialize industry as much as you will and still there will be more work for each of these kinds of people to do besides all the more supplying work for a third class of experts, the managers, the Henry Ford’s, the Henry J. Kaiser’s, the J. F. Lincoln’s, and others of lesser publicity and prominence. Expertness is important in all thesd realms. Selection is both positive and negative. When looking for traits that are rare-in consequence of which we pay well for them-we wish to include as many as possible of the desired traits in one man; we look ideally for the one man of all-men who most completely can fill the bill, the one man who includes in his make-up all the positive virtues. Of Tom, Dick, Joe, and Sally there are a myriad; hence they are paid chiefly for their time rather than for their pattern of abilities, and here we may seek only to exclude a certain few undesirable or nega-

Some Concepts of Job Families and Their Importance in Placement

THERE are, it is asserted, some 30,000 or more occupations. That is a large number. We could hardly construct that many aptitude tests, say, in a century. To any other proposal affecting all the occupations one would have to make a similar comment. The task is too big; it would not get done. There is great need, accordingly, for somehow reducing the number of &dquo;kinds&dquo; of occupations. Could one sort out, for example, a small number of type-occupations which would stand for or represent the lot of them? This hope is analogous with the corresponding dream of psychologists regarding human types. They hope to be able to type all humanity into a relatively few &dquo;unique personality profiles,&dquo; or patterns. Thus, though the people in a given type still would differ considerably amongst themselves, such differences might be thought of as relatively unimportant. The people of a given type, however, by definition would be singularly alike in, say, such matters as aptitudes, health, drives, wants, and satisfactions. They might still differ greatly in race, color, religion, height, weight, appearance and other

A Comparison, by Work-Limit and Time-Limit, of item Analysis Indices for Practical Test Construction

FOR some years now it has been our practice to include in the 150 items of any newly-devised later form of the Ohio State University Psychological Examination some forty items of high validity selected from the immediately preceding form. This means that, in general, with the appearance of each new form some 110 new items are printed for the first time and are given a bona fide printed tryout. One year after their first administration all 150 items of each form, since Form 18 inclusive, have been subjected to item analysis by the Adkins-Toops five-fold equal frequencied criterion method (Adkins & Toops, 1937), on populations of 820 up to nearly 4000, to permit the best items of each form to be chosen. Thus in general &dquo;the best forty items&dquo; over the years should improve in validity since highly valid newcomers are always challenging, and replacing, some of the (poorer) items of the old. Our statistical experience with the items of a new form, reduced to indices, then is recorded on 81/2 by 11 stiff, ruled, printed cardboard sheets, our Item Pedigree Form. One card represents our entire experience with a given item. Each time a new item is administered for the first time its indices