Madness and Method in Educational Testing

Abstract

Everyone talks about those educationists who, specializing in methods, completely ignore content. Such educationists must be a shy breed, seldom crawling out to expose themselves to the contumely they deserve, for they never seem to be identified by name. In sober fact, such education ists are nonexistent, because the professional edu cator is much too busy to be guilty of such a chimera as method without content. The profes sional educator is a practical person, concerned with the day-to-day operation of that demanding enterprise called public education. The details of management which eat up the educator's working week allow little time for such intellectual ab stractions as method devoid of content. Educa tional method, as such, could be developed, and has been developed, only in the leisurely atmos phere of the liberal arts colleges and the gradu ate schools of our universities. The psychologists who flourish in those academic oases must be given the blame or credit for educational method as a disembodied concept. Despite this noble parentage, educational method has known few happy days. Its godfath ers include many of the greatest names in psy chology, but there has been more glory in clinical psychology and in a narrowly academic experi mentation than in the application of psychology to education. Almost three-fourths of a century separates the Principles of Psychology of Wil liam James from the very similar practical educa tional interest of B. F. Skinner. To summarize rather brutally a universe of experimentation, one may state three basic princi ples of educational method: Principle 1: Learning is an active process. Principle 2: Right responses are learned most rapidly if there is immediate knowledge of success ; this principle is called rein forcement. Principle 3: Wrong responses are eliminated most rapidly if there is immediate knowledge of error; this is called feedback. The term feedback was no doubt borrowed from electronics for prestige reasons. Its applica tion to the process of learning is, however, tech nically accurate. Cybernetics has been compelled to pay grudging admiration to the self-correcting efficiency of the human brain. The classical psychological experiment on re inforcement and feedback was performed by Thorndike and reported in 1927. Thorndike had a group of seven blindfolded subjects try to draw a line 4 inches long. In 400 trials they were given no indication of whether or not they had drawn the line correctly, so it is hardly surprising that they showed no improvement. The errors aver aged about an inch at the beginning of this part of the experiment, and remained practically the same to the end. Then the subjects were given 25 more trials with an opportunity after each trial to open their eyes and check the line they had drawn. In only four trials they were able to re duce their error to -^ of an inch. One hardly knows whether to applaud this ex periment or scoff. If it seems somewhat ridicu lous to thus so soberly document the obvious, it is surely deplorable that this demonstration for which there has been much subsequent further proof has been completely wasted on teachers in general. And one must include the teachers of psychology. During the thirty years following Thorndike's experiment few efforts were made by psychologists or anyone else to improve classroom instruction through immediate knowledge of re sults. Pressey invented during the 1920's a num ber of forms of teaching machine; psychologists gave them a blank stare. In the 1,400 page Hand book of Experimental Psychology, published in 1951, Pressey's name is not indexed. American psychology caught up with Thorn dike and Pressey about ten years after World War II and as a result of it. During World War II, B. F. Skinner conducted experiments in pigeon guidance of missiles. The training of pigeons to