An Evaluation of Differences in Teaching Ability

Abstract

Of all the factors entering into the process of education, teaching ability has most stubbornly refused to submit itself to objective evaluation. Reasons for this are at least two in number. In the first place, we have not been able to agree upon the functions and importance of the classroom teacher, per se, either in general or in connection with instruction in par ticular fields at definite levels. Moreover, such evaluation may be rela tively unimportant since it has been said that is reason to doubt whether the differences between the very best and the very poorest teachers of a school system are half or even a quarter as important in determining the accomplishments of the pupils as are the differences in endowment as revealed by mental tests.1 The best we seem to have been able to do in this regard is to select a few more or less vague generalities on the basis of which we make our estimates of classroom efficiency. But there stands also a second obstacle, related to the first, in the path of an objective treat ment of this efficiency. Assuming that we have sharply defined the fea tures of the teaching act upon which we are to base our judgments, we still lack a meaningful numerical scale upon which to locate these judg ments. We may take any one of a number of teacher-training scales, and, using it, get for a particular teacher a total of, say, 155 points out of a possible total of 250. One hundred and fifty-five what ? No one knows. It would seem, then, that whatever method of rating is employed, the results must be capable of interpretation in terms of units already defined and generally understood. In the vocabulary of educational measure ment, no unit is so frequently used as mental age. Why not, then, use mental age?of students, to be sure?as a basis for a scale of teacher rating? Terman2 has stated the problem thus: . . .if inequalities of forty points in intelligence cannot be counterbalanced by reversed in equalities of teaching, how much intelligence quotient can be wiped out ? Is it thirty points, or twenty, or ten, or even less? For the purposes of the study described in the following pages, the problem has been restated to read: What is the mean increment of mental age necessary in order